Skip to main content

Full text of "Bulletin of the Museum of Comparative Zoology at Harvard College"

See other formats




HARVARD UNIVERSITY 

Library of the 

Museum of 

Comparative Zoology 



Bulletin of the Museum of Comparative Zoology 

AT HARVARD COLLEGE 

Vol. LXXXIX, No. 1 



A FOSSIL RIVER DOLPHIN FROM FLORIDA 



By Glover M. Allen 



With Three Plates 



CAMBRIDGE, MASS., U. S. A. 

PRINTED FOR THE MUSEUM 

October, 1941 



No. 1 — A Fossil Rirer Dolphin from Florida 
By Glover M. Allen 

In June of 1940 Professor Robert L. Nichols of Tufts College re- 
ceived a letter from Mr. Lewis H. Carter of the American Agricultural 
Chemical Company, stating that they were uncovering some verte- 
brate fossils in their Florida phosphate-rock mines. Curiously enough, 
these finds are not continuously made but occur at more or less hap- 
hazard and sometimes at quite widely separated periods of time. 

Knowing that the Museum of Comparative Zoology was particu- 
larly interested in such material, he brought the letter to Dr. P. E. 
Raymond, who called it to the attention of Dr. Thomas Barbour. 
Through Dr. Nichols a letter of introduction to the superintendent 
of the American Agricultural Chemical Company's mines, Mr. H. L. 
Hudson, was very kindly given Dr. Barbour by Mr. Carter. 

Owing to Dr. Barbour's subsequent illness he was unable to present 
this letter himself, but his associate. Dr. T. E. White, who was with 
him in Florida, received most courteous attention from Mr. Hudson, 
and through him introductions were given to the superintendents of 
other mines. Grateful acknowledgments are also due to Mr. W. N. 
Simpson, the company's engineer at Pierce. As a result of sundry 
visits to several mines in this area a number of extraordinarily in- 
teresting fossils from this productive early Pliocene locality have been 
acquired by the Museum, and one of these Dr. Barbour has entrusted 
to me to describe, as he suspected that it represented a new form. 

The specimen proves to be of unusual interest as the first certain 
record of a cetacean of the family Iniidae from eastern North America. 
Elsewhere, according to Dr. Remington Kellogg's account in 1928, 
fossil iniids are known from "a few imperfectly preserved fragments of 
skulls of Miocene and early Pliocene age," from two formations in 
Argentina and from an upper Miocene station near Rodeo, Cali- 
fornia. While the two living genera of this family, Inia and Lipotes, 
are called river dolphins on account of their habitat in the Amazons 
and the Yangtse River systems, respectively, it by no means follows 
that the fossil members of the family were confined to fresh water. 
Nevertheless, the name is convenient as a group designation. 

The Florida specimen apparently represents an undescribed genus 
and species, which may appropriately be dedicated to Mr. Hudson in 
recognition of his generosity and of his Uvely interest in the fossils of 
the phosphate formation. Sellards (1915) in his account of the pebble 
phosphates of Florida indicates that these "deposits were accumulated 



4 bulletin: museum of comparative zoology 

during either late Miocene or early Pliocene time." Some of the fossil 
vertebrates appear to be of fluviatile or estuarine origin, such for 
example as the iniid here discussed and the gavial Tomistoma. 

GoNiODELPHis new genus 

An iniid larger than the living Inia; beak very narrow, the tooth 
rows separated in the distal portion by a knife-like ridge, whence they 
suddenly diverge proximally to form an inverted Y. In the hinder 
part of the palate the two depressions between the (broken) vertical 
wing of the palatal and the midline, partially roofed over by the 
pterygoids when intact, form a pair of shallow gutters which are 
parallel instead of converging forward as in most small odontocetes; 
the floor of each gutter is smoothly continuous with the inner wall of 
the posterior nares as if it were in life open to the nasal passage. The 
latter slopes strongly upward and backward instead of being nearly 
vertical, forming an angle of some 30° with the plane of the palate. 
Posterior teeth with roots circular in section below the crown, but 
the tooth at the angle of the Y on each side triangular in section with 
the base of the triangle formed by the outer side. Derivation: gonia, 
angle, and del phis, dolphin, in reference to the wide angle of the tooth 
row. 

Type. Goniodelphis hudsoni new species. 

GoNiODELPHis HUDSONI new spccics 

Type. A portion of a cranium, No. 3920, Palaeont. Cat. Museum 
of Comparative Zoology, from phosphate deposit at Pierce, Florida. 
Collected by Mr. H. L. Hudson. 

Horizon. Probably early Pliocene. 

Description. The specimen consists of the basal part of the beak 
with palate, back to and including the nares; the proximal part of the 
tooth rows contains the roots of three teeth on the left side and eight 
on the right, distal to which are the broken alveoli for eight or nine 
additional pairs of teeth, narrowly spaced. The fragment indicates 
an animal about a third longer than the living Inia. The end of the 
narrow beak is missing but the part that is preserved represents pos- 
sibly half the original length. The lateral portions of the maxillaries 
and orbit are broken away as is also the ventral region behind the 
posterior nares and the entire occipital wall; at the posterior dorsal 
rim, however, is the clear outline of the supraoccipital. 



ALLEN: A FOSSIL RIVER DOLPHIN FROM FLORIDA 

Dorsal aspect: There is a marked asymmetry in the maxillary and 
intermaxillary bones, those of the left side are narrower basally (from 
median line to right margin of skull at level of pre-orbital notch, 71 
mm.; on left side, 47; width of right intermaxillary, 18; of left, 16 at 
same point). Dorsally the intermaxillaries taper forward to a com- 
bined width of 12 mm. at the broken tip against 34 at base; the two 
bones of opposite sides converge and unite at the level of the orbital 
notch, forming an elevated dorsal ridge with nearly vertical, slightly 
rounded sides, while proximally they diverge back to the level of the 
blowholes which they embrace. Anterior to the latter is a narrow, 
elevated and flattened area, about 130 mm. long. Below the inter- 
maxillary ridge, the outer sides of the maxillaries are deep and nearly 
vertical, and set off from the ridge by a shallow lengthwise groove. 
Most of the posterior part of the maxillaries is lost, but on the left 
side enough remains to show that, as in Inia, they turn up posteriorly 
to form a wide shelf with a similar gutter-like depression between its 
outer edge and the blowholes. The latter are of about ecjual size (some 
23 mm. each in transverse diameter). The nasal passage slopes 
strongly backward at an angle of about 30° with the plane of the 
palate. A shallow depression about 25 mm. long and deeper forward, 
is present at the hinder margin of each passage, marking the position 
of the squarish nasal bones. The summit of the skull was apparently 
less elevated than in Inia; its abraded outline is convex forward, 
brace-shaped posteriorly and about 32 mm. across its slightly trun- 
cated base. Immediately behind this summit (formed doubtless by 
the compressed frontals and parietals) is an emargination for the re- 
ception of the summit of the supraoccipital, convex forward on each 
side and concave medially, — brace-shaped, — with an extreme width 
of 55 mm. 

Lateral aspect: The depth of the rostral part of the maxillary and 
its solidity are striking features. At the level where the tooth rows 
diverge the depth of the maxilla is 39 mm., that of the overlying ridge 
formed by the intermaxillaries 19, a total vertical depth of 58 mm. 
At this point too, a low lateral ridge commences at the upper third of 
the maxillary, running back along the rising lateral edge of the bone, 
and parallel to the ascending wing. 

Ventral aspect: This view offers very distinctive characters. On 
the dextral side the orbito-maxillary projecting shelf is broken off 
apparently at its very base, leaving an indication of the bottom of the 
notch. From that point to the rim of the last alveolus of the tooth row 
is a space of 30 mm.; thence to the point where the alveolar rows of 



6 bulletin: museum of comparative zoology 

opposite sides meet is a distance of 110 mm. On the sinistral side the 
distance is sHghtly less, 100 mm., since there is one less tooth on that 
side. On the dextral side of the specimen there are roots of eight teeth 
in place forward from the penultimate socket, to the point where the 
two rows meet; on the sinistral side but three at the latter point. In 
advance of this the alveolar rows become parallel and so closely ap- 
proximated that only a narrow knife-like ridge intervenes between 
them. The alveolar walls are slightly broken away exteriorly so that 
it is difficult to make out the individual sockets of which there were 
apparently eight or nine similar in size to those of the proximal part 
of the row. Three striking characters here are: — (1) the extreme 
narrowness of the palate between the tooth rows that form the stem 
of the Y, and its sudden widening proximally so that the two alveolar 
rows forming the arms of the Y enclose between them an angle of 
about 50 degrees; (2) in the center of this enclosed triangular space is 
a narrow raised line of bone some 75 mm. long, on either side of which 
is a shallow lengthwise groove, 5 to 7 mm. wide, apparently corre- 
sponding to a similar ridge in Inia, formed by the vomer; (3) the tooth 
on each side at the point of divergence of the alveolar rows has the 
root still in place, and both are peculiar in being clearly triangular in 
section, with the longest side of the triangle external, and parallel with 
the maxillary border; the two other sides are shorter and subequal, 
forming the tip at the inner side. The next succeeding root on the 
dextral side is somewhat similar but less sharply angled, with the peak 
of the triangle external, while the corresponding tooth on the opposite 
side has a circular root and is less closely wedged against its neighbor. 
The remaining roots are circular or nearly so in section, one or two 
with slight evidence of a transversely oval form at the upper part of 
the root. 

The region of the palatal bones is again characteristic. In Lnia as in 
the delphinids, a thin blade-like wing extends ventrally from the lateral 
part of the palatal bone, cutting off a deep groove on each side of the 
median line in front of the posterior nares. In the delphinids the inner 
side of the pterygoids forms the anterior wall of the narial opening 
while the outer side folds over laterally to form a thin wall which 
unites with the palatal wing, thus completely enclosing a narrow 
fissure. In Inia this outer fold of the pterygoid is so short that it does 
not unite with the palatal wing, so that in the cleaned skull there is an 
open space between the two bones. In Inia, again, the two grooves 
enclosed by the palatal wings converge forward toward the mid-line 
of the skull. In the fossil skull here described the two grooves instead 



ALLEN: A FOSSIL RIVER DOLPHIN FROM FLORIDA 7 

are parallel to each other and relatively long (95 mm.), while their 
smooth surface is uninterruptedly continuous with the front wall of 
the nasal passage as if the latter had opened into the palatal fissure 
in life. 

Posterior to the nasal passage the roof of the brain case is seen from 
within. What appear to be the two parietals may be outlined at the 
posterior border, some 22 mm. long, with a combined transverse width 
of 60 mm. where they abutted against the supraoccipital. 

No part of the jaw was found, but evidently its rami were very 
closely approximated to form a long symphysis, then diverged ab- 
ruptly and widely near the base of the rostrum, in correspondence 
with the sudden divergence of the upper tooth rows at that point. 

Comparison with other iniids. In the table of fossil cetaceans ac- 
companying his paper on 'The History of Whales', Dr. Remington 
Kellogg (1928) lists but five genera of Iniidae then known. Four of 
these are from Miocene formations of Argentina and one from the 
Miocene of California. Of these, Anisodrlphis {A. brevirostris Rovereto) 
is based on a nearly complete jaw with long narrow syinphysis, but in 
which only two or three teeth occur posterior to the divarication of 
the mandibles. Furthermore, with the sixth or seventh of the series 
(counting from the posteriormost tooth), the teeth become widely 
spaced and those of the two sides alternate instead of being opposite. 
Apparently Rovereto's (1915) Sauroddphis acutirostris is rather similar. 

Ischyrorhynchus, as illustrated by True (1910, pis. 1-3) on the basis 
of a well-preserved skull which he describes under the name of Dio- 
chotichus vanbenedeni, shows a relatively broader rostrum, evenly 
tapering and with tooth rows widely separated, whereas in Goniodcl- 
phis the tooth rows are very closely approximated in front of the wide 
angle they make at the base of the beak. 

The genus Pontoplanodes of Ameghino was based on a fragment of a 
very narrow rostrum, with tooth rows closely approximated as in 
Goniodelphis, but with rather remarkable alveoli, long and narrow, 
for six opposite pairs of teeth quite different from those of the latter 
with their circular roots in section. The rest of the composite speci- 
men on which Burmeister based the name Saurodelphis argentinus is 
now believed to be identical with Ischyrorhynchus vanbenedeni. Its 
palatal aspect as figured by Abel (1909) shows very well the convergent 
palatal troughs and gradually tapering palate between the tooth rows. 

The fourth Argentinian genus, Proinia, with the single species P. 
patagonica True (1909), was based on fragments of a cranium from the 
Patagonian beds, Santa Cruz Territory. No comparison is possible, 



8 bulletin: museum of comparative zoology 

however, since the portions of the skull on which the genus and species 
were founded are practically the very ones missing in the Florida 
cranium. 

Finally Hesperocetus, known from only the type species, H. cali- 
fornicus True (1912), is a larger animal with teeth widely spaced even 
in the symphyseal region, and in such a way that those of the upper 
series were received into depressions between those of the lower. It 
was discovered in the Upper San Pablo beds near Rodeo, California. 

Of the two living genera, Inia and Lipotes, the former is perhaps 
the nearer related to the Florida animal as shown by its more ap- 
proximated tooth rows and the much narrower summit of the sup- 
raoccipital. Both, however, are widely different in details. 

LITERATURE CITED 
Abel, O. 

1909. Cetaceenstudien. II. Mitteilung: Der Schadel von Saurodelphis 
argentinus aus dem Pliozan Argentiniens. Sitzungsb. K. Akad. 
Wiss., Wien, math.-nat. Kl., sect. 1, 118, 255-272, pi. 1. 

Kellogg, Remington 

1928. The history of whales — their adaptation to life in the water. 
Quart. Review of Biol., 3, 29-76, 174-208, 24 figs. 

Miller, Gerrit S., Jr. 

1918. A new river-dolphin from China. Smithsonian Misc. Coll., 68, 
no. 9, 12 pp., 13 pis. 

Rovereto, Cayetano 

1915. Nuevas investigaciones sobre los delfines longirrostros del Mioceno 
del Parand (Republica Argentina). Anales Mus. Nac. de Hist. 
Nat., Buenos Aires, 27, 139-151, pis. 2-4. 

Sellards, E. H. 

1915. The pebble phosphates of Florida. 7th Ann. Rept. Florida State 
Geo!. Surv., pp. 25-116, 52 figs. 

True, Frederick W. 

1909. A new genus of fossil cetaceans from Santa Cruz Territory, Pata- 
gonia; and description of a mandible and vertebrae of Prosqualo- 
don. Smithsonian Misc. Coll., 52, 441-456, pis. 43-45. 

1910. Description of a skull and some vertebrae of the fossil cetacean 
Diochotichus vanbenedeni from Santa Cruz, Patagonia. Bull. 
Amer. Mus. Nat. Hist., 28, 19-32, pis. 1-5. 

1912. A fossil toothed cetacean from California, representing a new 
genus and species. Smithsonian Misc. Coll., 60, no. 11, 7 pp., 
2 pis. 



EXPLANATION OF PLATES 



PLATE 1 



Allen — A Fossil River Dolphin from Florida. 



PLATE 1 



Goniodelphis hudsoni. Dorsal view of the type specimen. The dotted out- 
line indicates the probable boundary of the maxillary shelf, x }4. Eugene N. 
Fischer, del. 



BULL. MUS. COMP. ZOOL. 



Allen. A Fossil River Dolphin from Florida. Plate 1 . 



£M 




PLATE 2 



Allen — A Fossil River Dolphin from Florida. 



PLATE 2 

Goniodelphis hudsoni. Ventral view of the type, showing (from front to 
back) the strongly approximated tooth rows, anteriorly, and their sudden wide 
divergence; the triangular section of the tooth roots at the angle of divergence; 
the median ridge anteriorly on the palate; the parallel grooves on the palatal 
bones; the brain case with outlines of the parietal bones at the posterior border. 
X H. E. N. Fischer, del. 



BULL. MUS. COMP. ZOOL. 



Allen. A Fossil River Dolphin from Florida. Plate 2. 







"V.^ '■■If: M^r■^ 

v-t> ■■■'.-'111 r'''r <■■'•'■ t?' ■■ t- 





PLATE 3 



Allen • — A Fossil River Dolphin from Florida. 



PLATE 3 

Goniodelphis hudsoni. Lateral view of the type specimen, showing the 
upturning maxillary wing, broken away, x I3. E. N. Fischer, del. 



BULL. MUS. CO MP. ZOOL. Allen. A Fossil River Dolphin from Florida. Plate 3. 




Bulletin of the Museum of Comparative Zoology 

AT HARVARD COLLEGE 

Vol. LXXXIX, No. 2 



STUDIES IX CUBAN BLATTIDAE (ORTHOPTERA) 



By Ashley Buell Gurney 

Bureau of Entomology and Plant Quarantine 
United States Department of Agriculture 



With Four Plates 



CAMBRIDGE, MASS., U.S.A. 

PRINTED FOR THE MUSEUM 

January, 1942 



No. 2 — Studies in Cuban Blattidae (Orfhoptera) 
Bt Ashley Buell Gurnet 

We sometimes think of the explorations of the early naturalists in 
the Neotropical Region, of men such as Humboldt, Bates, and Gund- 
lach, and of the rare pleasure that must have been theirs on discover- 
ing unusual plants and animals where white men had never been 
before. Undoubtedly, many strange creatures in tropical America still 
remain unknown to science, often in highl^^ inaccessible places, but 
only a few of us have the good fortune or initiative to seek them and 
thereby sense a joy of achievement akin to what the pioneer naturalists 
have known. One of the few North American entomologists to under- 
take seriously Neotropical exploration during recent years is P. J. 
Darlington of the ^luseum of Comparative Zoology, Harvard College, 
who has, by his collecting trips in the West Indies, greatly extended 
the frontier of our knowledge concerning the fauna of those islands, 
and who has found, especially in the higher mountains of the Greater 
Antilles, many strange, endemic insects. Dr. Darlington has brought 
back many remarkable Orthoptera, such as the katydids Polyancistrus 
and Poli/ancistroides (see Rehn, 1936, 1937b). 

The Blattidae taken by Dr. Darlington in Cuba in 1936 were 
loaned to the writer for study by Nathan Banks. As the study pro- 
gressed, and the importance of the collection became more apparent, 
the writer has been increasingly appreciative of the privilege of pre- 
paring this report. 

For courtesies extended to the writer while visiting the Academy 
of Natural Sciences of Philadelphia for the purpose of comparing 
types and identified material with specimens from the present collec- 
tion, thanks are due to J. A. G. Rehn, Curator of Entomology at that 
institution. As in connection with previous orthopterological studies, 
the privilege to consult the iVcademy Collection and the Hebard 
Collection located at the same institution is one for which the writer 
is particularly grateful. 

The collection consists of 142 specimens, comprising 16 genera and 
31 species. Of these, 8 species and 1 subspecies are here described as 
new. The holotypes and allotypes have been returned to the Museum 
of Comparative Zoology. One other new species, from the U. S. Na- 
tional Museum, is described. 

The majority of specimens were taken in Oriente Province in the 
mountains that characterize the eastern part of Cuba, and many were 
collected on the slopes or at the summit of Pico Turquino, the highest 



12 bulletin: museum of comparative zoology 

mountain in Cuba. This peak was not climbed until 1860 (see Taylor, 
1916), and, like other peaks which with it constituted an island 
archipelago during certain stages of Pleistocene time, has many 
species peculiar to it. Schuchert (1935) discussed the geological history 
of Cuba, and Darlington (1938), the theoretical aspects of different 
ways in which the animals of the Greater Antilles may have been 
established, with reference to the principal animal groups for which 
population statistics of genera and species are available. Bates (1935) 
and Barbour (1923) have briefly treated the topography and geology 
of Cuba, and Rehn and Hebard (1927) published a guide to the West 
Indian Blattidae which is important in all future work on the roach 
fauna of those islands. 

As a result of Dr. Darlington's collecting, series are now available 
of several species described by Rehn and Hebard in 1927 from one or 
very few specimens. The male of Aglaopicryx mira Rehn is here 
described for the first time. Especially in the case of Cariblattoides 
instigator R. & H. (see figs. 22, 23), the acquisition of a series shows 
that a great deal of variation occurs. Two of Bolivar's species de- 
scribed in 1888, Psevdosi/mplocc c.vcisa and Epilampra citbensis, have 
been "rediscovered." Modification of the generic limits of NeUpophy- 
gus has seemed advisable after studying a new species, banksi, which 
differs from the original generic diagnosis, but not sufficiently to 
warrant the proposal of a new generic name for it. Because the 1927 
monograph by Rehn and Hebard is well supplied with keys and illus- 
trations, it has not seemed necessary to prepare keys to species for all 
genera treated, but several keys have been given to supplement earlier 
ones, as in Eurycotis, to which several species have been added since 
1927. 

Some important observations have recently been made by Quadri 
(1940) on the male genitalia of Blatta oricntalis L. and Periplaneta 
aviericana (L.). He describes four parts of the inner genitalia, left 
and right dorsal penis valves and left and right ventral valves. In 
the third instar the left ventral valve is incompletely separated from 
the left dorsal valve,^ but is earlier differentiated and is distinguished 
in the adult on the basis of this interpretation. Following Snodgrass 
(1937), the writer uses "phallomere" rather than "penis valve." 
The Blattella type of genitalia (see Snodgrass, 1. c, p. 47) is not dis- 
cussed by Quadri. In valuable museum specimens of roaches it is not 
always advisable to make the dissections necessary for determination 

> It is stated in Qiiadri's paper (p. 145) that in this stage the left ventral penis valve is incom- 
pletely separated from the right dorsal penis valve, but reference to the illustration indicated by 
Quadri (pi. 4, fig. 30) suggests that "left dorsal valve" is intended instead of "right dorsal valve." 



GURNEY: CUBAN BLATTIDAE 13 

of the exact homologies of visible phallomeres. In addition to the 
phallomeres, and the supraanal and subgenital plates which have long 
been used taxonomically, the paraprocts often constitute important 
characters (see Gurney, 1939, for a treatment of the paraprocts of 
XestoblaUa). 

For the assistance of students, the following list of West Indian 
Blattidae described since 1927 is given. The islands of Tobago, 
Trinidad, Bonaire, Cura9ao, and Aruba are usually not considered 
West Indian in their affinities, and the species recorded from them 
are included simply for the sake of completeness. 

Aglaopicryx occulta Rehn 1932. Bermuda. 

Aglaopteryx mira Rehn 1932. Cuba. 

Aglaopteryx devia Rehn 1932. Puerto Rico. 

Aglaopteryx absimilis Gurney 1937. Puerto Rico. 

Aglaopteryx vegeta Rehn 1932. Jamaica. 

Cahita misella Rehn 1937. Tobago. 

Cariblatta adrena Rehn 1932. Hispaniola. 

Cariblatta faticana Rehn 1930. Cuba. 

Cariblatta spinicauda Hebard 1929. St. Vincent; Grenada. 

Cariblatta tobagensis Hebard 1929. Tobago. 

Eurycotis histrio Rehn 1937. Hispaniola. 

Eiirycotis improcera Rehn 1930. St. Croix. 

Eurycotis lixa Rehn 1930. Jamaica. 

Hormetica pustidata Hebard 1929. Bonaire (north of Venezuela). 

Nesomylacris fratercula Rehn 1930. Cuba. 

Pseudosymploce personata Rehn 1930. Puerto Rico. 

Styphoii bakeri Rehn, new genus and new species. Bonaire; Aruba; 
Curasao. 

It should be noted, in addition, that SibyUoblatta was proposed by 
Rehn, 1937, for Polyzosteria panesthoides Walker of Jamaica. 

Except for a few specimens so noted, all the material recorded in 
this paper is from Cuba. Some information, which is helpful in locat- 
ing localities mentioned on pin labels, is given by Darlington (1937) 
in a paper on carabid beetles taken on the same trip. 

Subfamily PSEUDOMOPINAE 

The Genus aglaopteryx Hebard 

Aglaopteryx Hebard, Mem. Amer. Ent. Soc, No. 2, p. 30, 1917, (genotype, 
A. gemma Hebard, by original designation). 

A synopsis of this genus by Gurney (1937) has recently appeared. 



14 bulletin: museum of comparative zoology 

Key to the Cuban Species of Aglaopteryx 

Small, length of pronotum about 3 mm.; interocular width less than 
three-fourths the distance between antennal sockets ; male subgenital 
plate with left style slender and acuminate (see Rehn, 1932, pi. 7, 
fig- 2) diaphana (Fabricius) 

Large, length of pronotum about 4.5 mm.; interocular width three- 
fourths or more the distance between antennal sockets; male sub- 
genital plate without well developed left style, as in fig. 24 

viira Rehn 

Aglaopteryx diaphana (Fabricius) 
Blatta diaphana Fabricius, Ent. Syst., vol. 2, p. 11, 1792. 

Material here recorded. 1 female, coast below Pico Turquino, Oriente 
Province, June 26-30, 1936 (P. J. Darlington) (M. C. Z.). 

This specimen agrees with the descriptive notes by Rehn (1932), 
who pointed out that the identity of this species had been largely mis- 
understood by previous workers. Measurements are as follows: 
Length of body 8.8 mm., of pronotum 3 mm., of tegmen 6.35 mm., of 
hind tibia 4.5 mm. ; width of pronotum 4.6 mm. 

Aglaopteryx mira Rehn 

Fig. 24 

Aglaopteryx mira Rehn, Trans. Amer. Ent. Soc, vol. 58, pp. 110-112, pi. 7, 
figs. 8-10, 1932. 

Material here recorded. 1 male, 5 females, 1 nymphal male, Pico 
Turquino (south side), Oriente Province, 3,000-5,000 feet, June 1936 
(P. J. Darlington) (M. C. Z. & U. S. N. M.). 

The species has hitherto been known only from three females. The 
present series of females agrees essentially with the original description. 
Differences in the distention of the abdominal segments account 
mainly for variation in body length, 11 to 15 mm., and in part for the 
fact that the tegmina of two specimens do not quite reach the apex 
of the abdomen. The two spots on the face between the antennal 
sockets figured by Rehn (1932, fig. 9) are not noticeable. The pro- 
notal length varies from 4.3 to 4.8 mm. 

Male. Essentially the same as female ; tegmina and wings surpassing 
apex of abdomen by about one-half of length of cercus; interocular 
width three-fourths the distance between antennal sockets. Supraanal 
plate transverse, apex barely emarginate. Subgenital plate (fig. 24) 



GURNEY: CUBAN BLATTIDAE 15 

with left posterior angle broadly rounded, margin noticeably decurved; 
a flap along right lateral margin abruptly bent in apical third, thus 
extending posteriorly and in a vertical plane; a blunt, swollen, bifid, 
median projection on posterior margin, the surface of left lobe heavily 
rugose; a triangular, acute projection to the left of the bifid process. 

Coloration. About as in female; disk of pronotum with side arms of 
an inverted anchor-shaped mark poorly developed, these arms con- 
nected by short marks to middle region of inner margin of each pro- 
notal bar. Subgenital plate pale brown, darker on lateral borders and 
on bifid apical process; supraanal plate pale. 

Measurements. Length of body 12 mm., of pronotum 4.5 mm., of 
tegmen 10.5 mm., of hind tibia 5.5 mm.; width of pronotum 6 mm. 

The male nymph is 11.5 mm. long. In addition to the w^ell-marked 
longitudinal pronotal bars, the disk of the pronotum has a conspicu- 
ous brown dot on each side, and along the middle line between them 
there is a weak longitudinal stripe which divides posteriorly. About 
eight brown dots occur on the posterior margin of the pronotum. 
The abdominal terga are spotted with brown, terga 4 and 5 more 
heavily so than the others. 

The Genus cariblatta Hebard 

Carihlatta Hebard, Trans. Amer. Ent. Soc, vol. 42, p. 147, 1916, [genotype, 
Blatta punctulata Beauvois (replaced by delicatula Guerin because of pre- 
occupation), by original designation]. 

This very characteristic West Indian genus of small, active roaches, 
represented by 23 species on the islands, is poorly represented in Cuba 
as regards number of species. Carihlatta faticana Rehn is known from 
a single female from Turquino Peak. 

Key to the Cuban Species and Subspecies of Cariblatta 

1. Pronotal pattern consisting of numerous dots of different sizes 

and more or less parenthesis-shaped marks; tegminal veins with- 
out conspicuous nodes 2 

Pronotal pattern essentially limited to two conspicuous dots on 
disk; most of costal veins and discoidal sectors of tegmen bear- 
ing on their dorsal surfaces one or more low rounded nodes. 

faticana Rehn 

2. Tinged with olivaceous; tegmina surpassing apex of abdomen, the 

latero-posterior angles of the pronotum being broadly rounded 
and the point of greatest pronotal width being about one-third 



IG bulletin: museum of comparative zoology 

of the pronotal length anterior to the posterior margin; wings 

fully developed delicatula (Guerin) 

Tinged with cinnamon; tegmina not reaching apex of abdomen 
except rarely, in short-tegmined specimens the latero-posterior 
angles of the pronotum being abruptly rounded and the point 
of greatest pronotal width being near posterior margin; wings 
vestigial lutea minima Hebard 



Cariblatta delicatula (Guerin) 

Blatta {Phyllodromia) delicatula Guerin, in La Sagra, Hist. Phys. Polit. Nat 
Cuba, pp. 346-347, 1857. 

Material here recorded. 1 male, 1 female, Maisi, Oriente Province, 
July 17, 1936; 2 males, mountains north of Imias, Oriente Province, 
3,000-4,000 feet, July 25-28, 1936; 2 males, Loma del Gato, Cobre 
Range, Oriente Province, about 3,000 feet, July 3-7, 1936; 1 male, 1 
female, Pico Turquino (south side), Oriente Province, 2,000-5,000 
feet, June 1936; 1 female, same, 1,500 feet, June 25, 1936; 4 males, 4 
females, coast below Pico Turquino, Oriente Province, June 26-30, 
1936; 1 male, Buenos Aires, Trinidad Mts., Santa Clara Province, 
2,500-3,500 feet. May 8-14, 1936; 2 females, Soledad (near Cien- 
fuegos), Santa Clara Province, April 1936 (P. J. Darlington) M. C. Z. 
& U. S. N. M.). 

The above material demonstrates the great variation found in this 
species which has already been discussed by Hebard (1916a, pp. 158- 
163) and Rehn and Hebard (1927, pp. 25-29). No species could better 
show the need for larger series in order to understand the amount of 
variation within a species or illustrate more clearly the fact that male 
genitalia, as well as other structural features, may vary widely. The 
posterior margin of the male subgenital plate varies from slightly 
concave to produced into a short median tongue. In one male (Buenos 
Aires) the projection is square, with a truncate apex; the projection in 
other specimens possessing a median tongue is bluntly rounded or 
triangular. The style at each latero-posterior angle varies from a tiny 
lobe armed at the base with a strong, curved spur and a small accom- 
panying spur to an elongate lobe armed with several spurs of various 
lengths along its outer and apical margins. Body size likewise varies 
greatly, the tegminal length ranging from 6.5 to 11 mm. in the male. 
There is an opportunity here for a fine contribution by some West 
Indian student interested in biology of insects who could rear this 



GURNEY: CUBAN BLATTIDAE 17 

species in quantity in order to tabulate the variation occurring in the 
progeny of known original stock. 

Cariblatta lutea minima Hebard 

Canblatta lutea minima Hebard, Trans. Amer. Ent. Soc, vol. 42, pj). 170-172, 
pi. 13, fig. 4, 1916. 

Material here recorded. 2 females, Soledad (near Cienfuegos), Santa 
Clara Province, April 1936; 1 female, same, May 1936 (P. J. Darling- 
ton) (M. C. Z. & U. S. N. M.). 

Readers are referred to Hebard (1917, pp. 50-56) and Rehn and 
Hebard (1927, p. 29) who have discussed this roach. The present sub- 
species occurs in Cuba and in the central and southern portions of 
Florida. C. lidea lutea (S. & Z.) is known from northern Florida to 
Cape Henry, Va., and westward to Natchez, Miss. The male sub- 
genital plate of minhna bears at each latero-posterior angle a short 
lobelike style armed with curved spurs. The posterior margin has a 
subquadrate median projection. It is unusual for a median projection 
of this shape to occur in delicatida, but the genitaha may not be de- 
pended upon to separate the two species. 

In addition to the distinguishing features noted in the key, the 
interocular distance of minima averages somewhat greater than that 
of delicatida. The differences in pronotal shape represent the progres- 
sive changes that occur in many species as an accompaniment of 
brachypterism. The only recorded specimen with long tegmina is 
the one from Baiios de San Vincente, Pinar del Rio Province, Cuba, 
mentioned by Rehn and Hebard (1. c, p. 29). No specimens with 
long wings are known. 

The Genus cariblattoides Rehn and Hebard 

Cariblattoides Eehn and Hebard, Bull. Amer. Mus. Nat. Hist., vol. 54, art. 1, 
p. 48, 1927 (genot,ype, C. suave Rehn and Hebard, by original designa- 
tion). 

Only the originally described species, suave and instigator, are 
known, but additional material suggests certain modifications in the 
key prepared in 1927. C. instigator was based on one male and five 
nymphs, and the present series of nine individuals, including two 
females, shows that much variation occurs. Three previously unre- 
corded males of suave, one from Rio Piedras, P. R., and two from 
Puerto Rico intercepted in quarantine at San Francisco, give further 
information about that species. 



18 bulletin: museum of comparative zoology 

Key to the Species of Cariblattoides 

Pronotum with two posteriorly diverging longitudinal dark bars on 
disk; tegmen with pale borders on both costal and anal margins; 
head with a distinct, transverse, dark interocular bar; male subgen- 
ital plate (figs. 22, 23) with both sides about equally developed. 
(Cuba) instigator Rehn and Hebard 

Pronotum with disk entirely dark, the dark area narrowing anteriorly; 
tegmen with pale costal border, anal margin entirely dark; inter- 
ocular area often darkened, but not in a sharply defined bar; male 
subgenital plate (Rehn and Hebard, 1. c, pi. 3, fig. 4) with left 
side much more developed and produced posteriorly than right side. 
(Puerto Rico) suave Rehn and Hebard 



Cariblattoides instigator Rehn and Hebard 

Figs. 22, 23 

Cariblattoides instigator Rehn and Hebard, Bull. Amer. Mus. Nat. Hist., vol. 54: 
art. 1, pp. 52-54, pi. 3, figs. 6-9, 1927. 

Material here recorded. 1 male, mountains north of Imias, 3,000- 
4,000 feet, July 25-28, 1936; 1 male, 2 females. Gran Piedra Range 
(near Daiquiri, between Santiago and Guantanamo), 2,000-3,000 feet. 
May 30-31, 1936; 1 male, summit of Pico Turquino, 6,000 feet, June 
16-21, 1936; 1 male, Pico Turquino (south side), 3,000-5,000 feet, 
June 1936; 2 males, same, 1,500 feet, June 25, 1936; 1 male, coast 
below Pico Turquino, June 26-30, 1936. All Oriente Province (P. J. 
Darlington) (M. C. Z. & U. S. N. M.). 

In the present series the fifth (apical) segment of the maxillary 
palpus averages distinctly shorter than the fourth, but not so much so 
as in suave. The number of small spines on the ventro-anterior margin 
of the front femur ranges from 8 to 13 with an average of 10.8; the 
range is 9 to 10 in the 2 females. In the 3 herein recorded males of 
suave, the spine range is 7 to 9. The number of tegminal discoidal 
sectors is 5 to 7 and 5 to 6, respectively, in the males and females of 
instigator, 6 to 7 in suave. 

The most conspicuous variation of instigator occurs in the male 
subgenital plate. One specimen (mountains north of Imias) (fig. 23) 
has a slender projection of the posterior margin of the subgenital 
plate, the styles being specialized as groups of strong curved spurs 
arising from a non-sclerotized area at the apex of each latero-posterior 
extension of the subgenital plate. Four specimens are as in fig. 22. 



GURNEY: CUBAN BLATTIDAE 19 

In this type there is no median projection, but there is a deep emargina- 
tion, while the styles arise from an unsclerotized area that is formed 
on the same plan as in fig. 23 except for less specialization. There is a 
thin, partly transparent area at the base of the median apical emargina- 
tion. A specimen from the south side of Pico Turquino is about as 
illustrated by Rehn and Hebard (1. c, pi. 3, fig. 9) as regards the 
median projection and the general shape of the subgenital plate, but 
the styles are represented by clusters of spurs about as in fig. 22. The 
single male from the summit of Pico Turquino is much like fig. 23 
except that the styles converge so that they virtually meet near the 
tip of the median projection, curving so that an opening is left along 
each side of the median projection. This male likewise differs from 
the other specimens, perhaps in response to the high altitude, by being 
of darker general coloration (the venter of the abdomen is mostly 
brown except for pale margins and a few spots), by the tegmina not 
quite reaching the apex of the abdomen, by the wings being vestigial, 
and by the pronotum being more semicircular than transversely ellip- 
tical as a result of brachypterism. If this specimen actually represents 
a population restricted to high elevations, it probably is a distinct 
subspecies of instigator. Further collecting may supply information 
on this subject. In two males of the above series the supraanal plate 
is narrowly incised apically; the apex of the others is entire. In each 
of two males there is a right phallomere extruded, which is in the form 
of a slender, curved, and somewhat twisted hook, recurved apically 
and armed with two sharp teeth. Tegminal length of the normal males 
varies from 9.5 to 11 mm. Measurements of the short-winged male 
from the summit of Turquino follow: Length of body 7 mm., of pro- 
notum 2.4 mm., of tegmen 5 mm., of hind tibia 3 mm.; width of pro- 
notum 3.5 mm. Females agree essentially with males in color and 
structure. The interocular space is not noticeably wider in the female. 
The female supraanal plate is transverse, weakly bilobed at the apex, 
and the apical margin has sparsely distributed, long, slender, spinelike 
setae. The subgenital plate is simple, scooplike, extending about half 
its length beyond the supraanal plate; it is pale buff, darker along 
the lateral margins. Measurements of a female are as follows: Length 
of body 8.2 mm., of pronotum 2.5 mm., of tegmen 10.8 mm., of hind 
tibia 3.9 mm. ; width of pronotum 3.4 mm. 

The Genus neoblattella Shelford 

Neoblattella Shelford, Ent. Mo. Mag., vol. 47, p. 155, 1911 (genotype, Blatta 
adspersicollis Stal, by original designation). 



20 bitlletin: museum of comparative zoology 

Few American genera of Blattidae are comparable to Neohlattella in 
the rich number of species, but only two species are known from Cuba, 
one of them here described as new. 

Key to the Cuban Species of Neohlattella 

Posterior margin of male subgenital plate (fig. 32) prominently incised 

and with a small spine each side of emargination 

guanayara, new species 
Posterior margin of male subgenital plate (fig. 38) not incised and 

without spines vatia Rehn and Hebard 

Neoblattella vatia Rehn and Hebard 

Fig. 38 

NcobldlUUa Pdtia Eehn and Hebard, Bull. Amer. Mus. Nat. Hist., vol. 54, art. 1, 
pp. 63-66, pi. 5, figs. 1-6, 1927. 

Material here recorded. 1 male, Upper Ovando River (the mouth of 
which is just south of Cape Maisi), Oriente Province, 1,000-2,000 feet, 
July 17-20, 1936; 1 female, Loma del Gato, Cobre Range, Oriente 
Province, about 3,000 feet, July 3-7, 1936; 1 male, Pico Turquino, 
Oriente Province, 5,000-6,000 feet, June 1936; 1 male, same, 6,000 feet 
(summit), June 16-21, 1936; 4 males, 2 females, same, 3,000-5,000 feet 
(south side), June 1936; 1 male, same, 1,500 feet, June 25, 1936. 

The origin of the holotype is unknown, beyond that it is from Cuba, 
but the other original material is from Pinar del Rio and Havana Prov- 
inces. The apex of the male subgenital plate is not produced in a 
swollen protuberance nearly so much in the types as in the present 
series (fig. 38), but all evidently represent the same species. Within the 
present series some variation occurs. Further collections may show 
that the population in eastern Cuba is a distinct subspecies. The 
apical margin of the male subgenital plate separates vatia from guana- 
yara (figs. 32, 38). The tegmina of two males (5,000-6,000 feet, and 
summit of Turquino) are noticeably shorter than those of the other 
specimens here recorded. The tegmen of the male from the summit 
of Turquino is 10 mm. long; that of the male taken at 1,500 feet is 17 
mm. long. 

Neoblattella guanayara, new species 

Fig. 32 

Male (holotype). Differing from mtia especially in subgenital plate. 
Head with interocular space narrower than distance between antennal 



GURNEY: CUBAN BLATTIDAE 21 

sockets (as 2.7 to 3.6) ; pronotum slightly more transverse, rather more 
delicate than in vatia; ulnar vein of wing with four complete rami, 
each of the two most apical ones twice furcate; axillary vein with two 
branches, the basal one furcate near its base and each fork furcate 
about midw ay of its length ; apical margin of supraanal plate narrowly 
concave, the margin of concavity appearing as two obtuse angles (this 
condition occurs in some males of vatia, in others the margin is entire). 
Subgenital plate with a large hooked style at each postero-lateral angle, 
as in fig. 32 in ventro-posterior view; a small curved spine near base of 
each style; apical margin weakly obtuse-angulate (in ventral view), 
prominently incised; a small curved spine each side of emargination; 
a median, slightly oblique, longitudinal ridge in posterior half of plate, 
culminating in a protuberance extending ventrad of emargination; 
lateral portions of plate abruptly curved dorsad and situated nearly 
vertically, serving as bases for respective styles. 

Coloration. In general, light straw-colored, much as in vatia. Eyes 
black; interocellar area light brown with a pale central spot enclosed 
with dark; antenna brown, pale at base; pronotum, tegmina, wings, 
legs, and abdomen as in vatia; supraanal and subgenital plates light 
brown; cercus whitish yellow above, margined laterally wnth brown, 
ventral surface of segments pale, blotched basally and margined 
laterally with brown. 

Measurements. Length of body 11.5 mm., of pronotum 3.5 mm., of 
tegmen 15 mm., of hind tibia 6.5 mm.; width of pronotum 4.7 mm. 

Female (allotype). Interocular area slightly wider than in male, 
in respect to distance between antennal sockets as 3 to 3.8; supraanal 
plate with a V-shaped emargination as in vatia; subgenital plate simple, 
scooplike, extending somewhat beyond supraanal plate. 

Coloration. As in male; supraanal plate brown on base, apical half 
pale ; subgenital plate dark brown, darkened on margins, pale on central 
part of disk, a pale basal blotch with dark-brown dot on each lateral 
border. 

Measurements. Length of body lO.S mm., of pronotum 3.6 mm., of 
tegmen 14.7 mm., of hind tibia 6.2 mm.; width of pronotum 4.7 mm. 

Type locality. Buenos Aires (about 17 miles northwest of Trinidad), 
Trinidad Mts., Santa Clara Province, Cuba. 

Type. Museum of Comparative Zoology. 

The type and allotype were taken by P. J. Darlington at the type 
locality, 2,500-3,500 feet. May 8-14, 1936. Buenos Aires is near the 
headwaters of the Rio Guanayara and the Rio San Juan, from the 
former of which the specific name is taken. 



22 bulletin: museum of comparative zoology 



The Genus pseudosymploce Rehn and Hebard 

Pseudosymploce Rehn and Hebard, Bull. Amer. Mus. Nat. Hist., vol. 54, art 1, 
p. 103, 1927, (genotype, P. schistopyga Rehn and Hebard, by original 
designation). 

Since 1927, Rehn (1930) has described personata from Puerto Rico, 
and now excisa (Bol.) has been taken in Cuba, doubling the number of 
species reported from x\merican collections when the revision of West 
Indian roaches was written. Both 'personata and excisa have four com- 
plete rami of the ulnar vein of the wing, and the tegminal discoidal 
sectors of c.rcisa are weakly oblique, thus modifying slightly the original 
generic description, which states that the complete rami are three in 
number and the discoidal sectors are longitudinal. The descriptions 
and illustrations of the 1927 revision should be consulted in conjunction 
with the key below. 

Key to the Species of Pseudosymploce 

1. Noticeably bicolored, anterior and lateral margins of pronotum 

and costal margin of tegmen yellow in contrast to dark-brown 
basal color; apex of female supraanal plate entire. (Discoidal 
sectors of tegmen about 12 in number.) (Puerto Rico) 

personata Rehn 

General color uniform, either solid blackish brown or pale tinged 

with auburn or rufous; apex of female supraanal plate bilobate 

or emarginate 2 

2. Large, blackish brown, tegminal length about 20 mm. or more; 

head entirely black or blackish brown except clypeus, which is 
pale brown; discoidal sectors of tegmen about 12. (Cuba) 

excisa (Bolivar) 

Smaller, reddish brown to pale, tegminal length about 14-17.5 mm. ; 

head not nearly so uniformly dark as in excisa; discoidal sectors 

of tegmen about 8 3 

3. Posterior margin of male supraanal plate with median emargina- 

tion much broader than length of posterior processes. (Jamaica) 

schistopyga Rehn and Hebard 
Posterior margin of male supraanal plate with median emargination 
no broader than length of posterior processes. (Hispaniola) 

elongata (Beauvois) 

Pseudosymploce excisa (Bolivar) 
Ischnoptera excisa Bolivar, Mem. Soc. Zool. France, vol. 1, p. 124, 1888. 



GURNEY: CUBAN BLATTIDAE 23 

Material here recorded. 1 female, mountains north of Imias, Oriente 
Province. 3,000-4,000 feet, Julv 25-28, 1936 (P. J. Darlington) (M. 

This specimen agrees very closely with Bolivar's description and, 
except as already noted, fits the original generic description perfectly. 
According to Gundlach (1890-91, p. 303), Bolivar's material came from 
Yateras, Oriente Province; this is scarcely 30 miles from Imias. The 
rediscovery of excisa, unknown since its description more than a half 
century ago, is of genuine interest. 

Female. Large for genus, rather robust; surface smooth and glossy. 
Narrowest width of interocular space slightly wider (about as 5.5 to 5) 
than distance between antennal sockets; eye extending far ventrally, 
to a point fully one-half distance between level of ventral margin of 
antennal socket and lateral articulation of mandible. Maxillary palpus 
with segments 5 and 4 subequal in length, each slightly shorter than 
segment 3; segments slender, apical one very shallowly, triangularly 
hatchet shaped in lateral view. Pronotum as in elongata. Tegmen ex- 
ceeding tips of cerci by distance about equal to pronptal length; num- 
ber of costal veins about 20; discoidal vein forked about niidlength of 
tegmen, the posterior branch forked one-fifth the distance from origin 
of vein to apex; discoidal sectors 12 (inclusive of ulnar vein and exclu- 
sive of posterior fork of discoidal vein), weakly oblique. Wing reaching 
about to apex of tegmen; costal veins 14, including mediastine vein, 
which extends across bases of about 6 costal veins, forming about 7 
closed cells anterior to middle third of discoidal vein; median vein 
simple ; ulnar vein with 4 complete and 3 incomplete rami ; intercalated 
triangle about one-fifth as long as length of anal sulcus; axillary vein 
with 2 simple branches, in addition to main stem; radiate veins 13. 
All femora much compressed; ventro-anterior margin of front femur 
with 8 large spines, the apical 2 considerably shorter than the others, 
then 8 small, imiform, setalike spines; other legs about as in elongata. 
Supraanal plate transverse, moderately produced, the apex with deep, 
broadly V-shaped emargination; subgenital plate simple, bluntly 
rounded at apex, noticeably surpassing supraanal plate posteriorly. 

Coloration. General color blackish brown; tegmen tinged with 
metallic luster; pronotum mostly dark reddish brown except for a 
darker, wide, posterior border; eyes blackish; occiput and face dark 
reddish brown; clypeus paler; antennal sockets and ocellar spots light 
yellow; antenna light brown, darker at base; maxillary palpus brown, 
apical segment pale in apical half; legs brown; wing hyaline, fuscous 
along anterior border and in apical third of anterior field; abdomen 



24 bulletin: museum of comparative zoology 

dark brown, lighter in middle region of venter; cerci and supraanal 
plate dark. 

Measurements. Length of body 17 mm., of pronotum 5.4 mm., of 
tegmen 21.5 mm., of hind tibia 8.5 mm.; width of pronotum 6.4 mm.; 
greatest width of tegmen 6.5 mm. 

The Genus ischnoptera Burmeister 

Ischno-ptera Burmeister, Handb. Ent., Bd. 2, Abt. 2 (1st half), p. 500, 1838, 
(genotype, Ischnoptera morio Burmeister, designated by Kirby, 1906). 

Many species of Ischnoptera inhabit Central and South America, but 
only three have been found in Cuba. Elsewhere in the West Indies, 
podoces R. & H. and orcocharcs R. & H. occur in Jamaica, rufa rufa 
(Deg.) is widely distributed in the Greater and Lesser Antilles, and 
ligula R. & H. Avas recorded from Hispaniola by Rehn and Hebard 
(1927, p. 119) on the basis of two females. 

Key io the Cuhan Species and Subspecies of Ischnoptera 

1. Large, chestnut colored, width of pronotum usually at least 5 mm., 

occasionally about 4 mm. ; apex of male supraanal plate (Hebard, 

1916b, pi. 16, fig. 3) entire rufa rufa (Degeer) 

Smaller, usually less reddish, width of pronotum about 2.5 to 3.5 
mm.; apex of male supraanal plate (fig. 25, 31) conspicuously 
emarginate 2 

2. Male supraanal plate (fig. 31) very deeply incised, appearing asym- 

metrically bilobed darlin.gtoni, new species 

Male supraanal plate very difl^erently emarginate from above, about 
as in fig. 25 . ligula Rehn and Hebard 3 

3. Robust; length of tegmen about 10-11 mm. ; middle part of apex of 

male subgenital plate recurved dorsally, bearing a very special- 
ized right style (fig. 26); apical emargination of male supraanal 
plate averaging larger (fig. 25) .... ligula collina, new subspecies 
Less robust; length of tegmen about 8.5 mm.; apex of male sub- 
genital plate not recurved dorsally, right style less specialized 
(Rehn and Hebard, 1. c, pi. 9, fig. 7); apical emargination of male 
supraanal plate averaging smaller than above 

ligula ligula Rehn and Hebard 

Ischnoptera rufa rufa (Degeer) 
Blatta rufa Degeer, Mem. Hist. Ins., vol. 3, p. 539, pi. 44, fig. 7, 1773. 



gurney: clban blattidae 25 

Material here recorded. 1 male, intercepted at Boston, Mass., in 
quarantine from Cuba, 1932; 1 male, intercepted at Norfolk, Va., in 
quarantine from Cuba, April 6, 1937 (U. S. N. M.). 

By coincidence Brunner (1865, p. 131) used the same specific name in 
describing his synonymous Ischnoptera rufa from Brazil. Moreover, 
Brunner omitted reference to Degeer's rufa in his review (1. c, p. 30) 
of the work of that author. Kirby (1904) omitted Degeer's name also, 
but later included it (1910, p. 506) as a species of uncertain position. 
The synonymy of rufa Brunner was published by Shelford (1908, p. 8). 
This species should not be confused with Blatta rufa (Tepper) of 
Australia.^ 

This is the first Cuban record of rufa rufa, which elsewhere occurs 
throughout the \Yest Indies and on the mainland from Nicaragua 
southward over much of northern South America. Hebard (1916b, 
1920) has illustrated the male genitalia. The subspecies rufa debilis 
Hebard occurs "in the higher country of Costa Rica," and rufa ocd- 
dentalis Saussure from Panama northward. 

Ischnoptera darlingtoni, new species 
Figs. 27-31 

This species is most closely related to ligula and oreochares, from 
both of which it differs in the male genital structures. 

Male (holotype). General form slender; prnontum and tegmina 
polished. Head rather narrow in frontal view; interocular area nar- 
rower than distance between antennal sockets (as 2.5 to 3.2); face 
moderately inflated ; segment 5 of maxillary palpus elongate-lanceolate 
in lateral view, about one and one-third times as long as segment 4; 
pronotum as in ligula. Tegmen exceeding apex of abdomen by distance 
of 5 mm.; costal veins (including mediastine vein) 21 in number; dis- 
coidal sectors as in ligula, 9 to 11 in number. Wing with about 19 
costal veins ; ulnar vein with 2 complete and 5 incomplete rami ; inter- 
calated triangle small, about one-fifth the length of anal sulcus; axil- 
lary vein with 2 branches; about 10 radiate veins. Ventro-anterior 
margin of front femur with 1 strong spine and several very slender, 
long ones along basal portion; about 12 setalike spines in apical half; 
3 large spines at extremity as in ligula. Other legs about as in ligula. 

Supraanal plate (fig. 31) deeply incised; right lobe deeply concave 
ventrally, with numerous small spines closely set along margins ven- 

1 Described bv Tepper (1893, p. 101) as Periplaneta rufa. Transfer of the species to Blatta by 
Kirby (1904, p. 139) and Shelford (1910, p. 16) made this a secondary homonym. Accordingly, 
Blatta tepperatm, new name, is here proposed for the Australian species. 



26 bulletin: museum of comparative zoology 

trally; left lobe somewhat concave ventrally, a group of long, strong 
spines directly obliquely on inner margin, another group of spines at 
left near base ventrally. Subgenital plate appearing as in fig. 30 in 
dorso-posterior view, a median dorsally recurved flap bearing highly 
specialized right style (fig. 27) ; left style slender, acute, bearing three 
lateral spines. Left and right paraprocts as illustrated (figs. 28, 29), 
respectively. 

Coloration. General color light brown; interocular area and dorsal 
part of face orange yellow; ventral part of face and palpi pale; ocellar 
spots white, tinged with pinkish; antenna buff, basal three segments 
pale. Pronotum bordered laterally and anteriorly wath pale yellow; 
disk orange yellow, bordered with fuscous in a wide posterior border 
and posterior to the anterior pale border. Tegmen light brown, washed 
with whitish along costal margin indistinctly toward apex; apex of 
tegmen and posterior half of discoidal field grading into hyaline. Wing 
membrane colorless; costal veins clouded in middle third of wing; 
veins light brown. Legs pale; tarsi and spurs somewhat darker. Abdo- 
men pale, darker toward apex ventrally and on posterior portions of 
terga. 

Measurements. Length of body 9 mm., of pronotum 2.7 mm., of 
tegmen 12.5 mm., of hind tibia 4 mm.; width of pronotum 3.5 mm. 

Female (allotype). General form as in male; somewhat larger and 
more robust. Interocular space slightly wider than distance between 
antennal sockets (about as 4 is to 3.7). Tegmen with about 26 costal 
veins; 11 discoidal sectors. Wing with 2 complete and 7 to 8 incomplete 
rami of ulnar vein. Front femur with 2 large spines basad of setalike 
spines on ventro-anterior margin. Supraanal plate transverse; apex 
blunt, about a right angle. Subgenital plate equal in length to supra- 
anal plate, apical margin broadly and evenly rounded. 

Coloration. About as in male, but slightly more tinged with orange 
yellow. Ocellar spots white. 

Measurements. Length of body 9.5 mm., of pronotum 2,7 mm., of 
tegmen 12.5 mm., of hind tibia 4 mm.; width of pronotum 3.4 mm. 

In addition to the type and allotype just described, there is 1 male 
paratype. The armature of the right style varies slightly from that of 
the type. There are about 20 costal veins in the tegmen, and the com- 
plete and incomplete rami of the ulnar vein of the 2 wings are, respec- 
tively, 2 and 6, and 2 and 5. There are 2 large spines on the ventro- 
anterior margin of each front femur. The coloration tends more toward 
fuscous and less toward orange yellow than in the type. 

Type locality. Mountains north of Imias, Oriente Province, Cuba. 



GURNEY: CUBAN BLATTIDAE 27 

Type. Museum of Comparative Zoology. 

Paratype. U. S. National Museum, No. 54826. 

The holotype was taken at an altitude of 3,000-4,000 feet, July 25- 
28, 1936, by P. J. Darlington, in whose honor the species is named. The 
allotype is from Pico Turquino (south side), Oriente Province, 3,000- 
5,000 feet, June 1936. The paratype is from Loma del Gato, Cobre 
Range, Oriente Province, about 3,000 feet, July 3-7, 1936. The allo- 
type and paratype also were taken by Dr. Darlington. 

IscHNOPTERA LIGULA COLLINA, new subspecies 
Figs. 25, 26 

This roach differs from ligula ligida as noted in the key, by the 
larger number of areolets between the discoidal and median veins of the 
wing, by a slightly wider interocular space, and by a smaller average 
number of large spines along the basal half of the ventro-anterior 
margin of the front femur. Typical ligula occurs at low altitudes in 
both western and eastern Cuba, and the present form appears to be a 
subspecies occurring at higher elevations (the name collina means 
"one who dwells among the hills"). 

Male (holotype). Interocular width less than distance between an- 
tennal sockets (about as 2.5 to 3.4) (a male paratype of ligula ligula is 
as 2.2 to 3.5); ventro-anterior marign of front femur with 1 large spine 
basad of a row of fine setalike spines; wing with 12 subquadrate areo- 
lets between discoidal and median veins (8 to 10 in paratypes of ligula 
ligula). Supraanal plate (fig. 25) somewhat asymmetrical at base; 
lateral portions recurved ventrally; apical emargination, accentuated 
by curvature of lateral portions, deeply rounded; numerous small setae 
lining apical portion within; sparse elongate setae as illustrated. Sub- 
genital plate, as in fig. 26 in dorso-posterior view, with apical margin 
recurved dorsally; left style slender, gently curved, acute, bearing a 
small lateral spine; right style sulcate dorsally in basal half, armed with 
3 heavy spines. Left paraproct bearing a slender arm, with an acute 
apical spine ; right phallomere slender, curving dorsally, then ventrally 
and to the left near base of right style, bearing a curved apical spine. 

Coloration. Essentially as in ligula ligula. 

Measurements. Length of body 8.5 mm., of pronotum 2.3 mm., of 
tegmen 11 mm., of hind tibia 3.8 mm.; width of pronotum 3.1 mm. 

Female (allotype). Much like male; interocular space subequal to 
distance between antennal sockets; about 15 areolets between discoidal 
and median wing veins; ulnar vein with 2 complete and 4 incomplete 



28 bulletin: museum of comparative zoology 

rami of ulnar vein. Supraanal plate with lateral margins broadly 
rounded, apex shallowly and narrowly emarginate; subgenital plate 
with apex broadly rounded and briefly reflexed dorsally. 

Coloration. Essentially as in male. 

Measurements. Length of body 7.5 mm., of pronotum 3 mm., of 
tegmen 10.5 mm., of hind tibia 3.5 mm.; width of pronotum 3.2 mm. 

In addition to the type and allotype, there are 2 male and 2 female 
paratypes. One female has 2 large spines on the ventro-anterior mar- 
gin of each front femur; the second female has 1 spine on 1 femur and 
2 on the other. A spread wing of 1 male paratype has about 14 areolets 
between the discoidal and median veins of the wing, and 1 female has 
about 18 areolets; the tegmen of the latter specimen measures 11.5 
mm., and the supraanal plate is entire apically. 

Type localiii/. Loma del Gato, Cobre Range, Oriente Province, Cuba. 

Type. Museum of Comparative Zoology. 

Paratypes. U. S. National Museum, No. 54827, and Museum of 
Comparative Zoology. 

The holotype was collected at an elevation of about 3,000 feet, July 
3-7, 1936. The allotype, one male paratype, and one female paratype 
bear the same data as the type. One male paratype is from the Gran 
Piedra Range (near Daiquiri), Oriente Province, 2,000-3,000 feet. 
May 30-31, 1936. One female paratype is from Buenos Aires, Trini- 
dad Mts., Santa Clara Province, 2,500-3,500 feet. May 8-14, 1936. 
i\ll the material was collected by P. J. Darlington. The two paratypes 
from the type locality are deposited in the U. S. National Museum. 

The Genus nelipophygus Rehn and Hebard 

Nelipophygui; Rehn and Hebard, Bull. Amer. Mus. Nat. Hist., vol. 54, art. 1, 
p. 122, 1927 (genotype, A^ rainsdeni R. & H., by original designation and 
by monotypy). 

The material collected by Dr. Darlington in Oriente Province of this 
unusual Cuban genus adds to our knowledge of the variation occurring 
in ravisdeni and includes a new species the characters of which necessi- 
tate two modifications in the original generic diagnosis. The tegmen 
of banksi IS truncate and weakly concave and oblique at the apex 
(fig. 35), rather than arcuate-truncate with the greatest length on the 
inner margin as in ramsdeni. The arolia are small but distinct (fig. 36) 
instead of absent as in ramsdeni. 

The close similarity of the two species in other important characters 
and in habitus, together with their occurrence in the same part of 



gurnet: CUBAN BLATTIDAE 29 

Oriente Province, indicates clearly that they belong to the same 
natural group. xVccordingly, the limits of Nelipophygus are modified 
to include banlxsi. In the key to West Indian genera of Pseudomopinae 
(Rehn and Hebard, 1927, pp. 5-7) ,banksi ends blindly near Ischnoptera 
and Xclipophi/gus, but the substitution of "Arolia absent or much re- 
duced in size" for "Arolia absent" eliminates the difficulty. Neither 
species runs directly through couplet 1 of that key, since their vestigial 
wings do not provide the characters there required, but by elimination 
the student will recognize them as members of the group Ischnopterae 
and then arrive at Nelipophygus by the key characters. 

Nelipophygus ramsdeni Rehn and Hebard 

Nelipophygus ramsdeni Rehn and Hebard, Bull. Amer. Mus. Nat. Hist., vol. 54, 
art. 1 pp. 123-126, pi. 9, figs. 8-13, pi. 10, figs. 1-3, 1927. 

Material here recorded. 1 female, Upper Ovando River (the mouth 
of which is just south of Cape Maisi), Oriente Province, 1,000-2,000 
feet, July 17-20, 1936 (P. J. Darlington) (M. C. Z.) 

This specimen has been compared with the types at Philadelphia and 
found to difter from the female allotype in smaller size and a lighter 
shade of brown. These differences may represent individual variation 
or reflect the influence of environment. Measurements are as follows: 
Length of body 15 mm., of pronotum 5.25 mm., of tegmen 5.7 mm., 
of hind tibia 6.4 mm.; width of pronotum 6.6 mm., of tegmen 4.6 mm. 

Nelipophygus banksi, new species 
Figs. 34-37 

This distinct species differs from ramsdeni in the small arolia, differ- 
ently shaped tegmina, more expanded hind tibiae, less elongate pro- 
jection on the posterior margin of metanotum, and in differences in the 
male subgenital plate. 

Male (holotype). General body form as in ramsdeni; dorsal surface 
moderately shining. Interocular space much wider than distance be- 
tween antennal sockets (as 10 to 6.6) ; antennae, eyes, and palpi as in 
ramsdeni. Pronotum with sides decurved hoodlike anteriorly, con- 
cealing head from above; a few punctations; shape as in ramsdeni. 
Tegmen (fig. 35) subquadrate, greatest length on costal margin ; latero- 
posterior angle reaching midway on tergum 2; inner posterior angle 
falling short of posterior margin of metanotum; apical margin weakly 
oblique, slightly concave; venation as illustrated. Wing a vestigial 
pad, reaching to posterior margin of metanotum, concealed by tegmen 



30 bulletin: museum of comparative zoology 

in dorsal view. Posterior margin of metanotum with blunt, obtuse- 
angled projection (not long and acutely spinelike as in ramsdeni.) 
Front and middle legs like those of ramsdeni except for presence of 
small arolia; ventral margins of hind femur with six to eight sturdy 
spurs; hind tibia noticeably more flattened and expanded than in 
ramsdeni, a broad, shall owly concave area on the surface of apical 
third adjoining external margin and extending two-thirds width of 
tibia, not grooved or so definitely limited in area as in certain species of 
Eurycotis (fig. 4) ; hind tarsus as in ramsdeni except for arolium (fig. 36). 
Abdomen shaped as in ramsdeni; posterior half of dorsal surface with 
conspicuous, recumbent, sparsely distributed setae; supraanal plate 
about as in ramsdeni , slightly emarginate apically; a left phallomere 
large, hooked, similar to that of ramsdeni (see Rehn and Hebard, 1. c, 
pi. 10, fig. 1); a right phallomere sharp, elongate, spinelike, projecting. 
Subgenital plate (fig. 34) with right lateral portion abruptly curved 
dorsally in a nearly vertical plane; left lateral portion less abruptly 
curved dorsally; apical margin with narrow, ventrally decurved flap; 
disk of subgenital plate broadly concave in posterior half; styles borne 
near base of apical flap ; left style slightly tapering, extending ventrad 
of plate apically; right style broadly ovate, extending about on plane 
with plate, a prominent oblique ridge on ventral surface. Cerci as 
in ramsdeni. 

Coloration. General color dark brown, with reddish tinge; eyes 
black; general color of face and occiput light reddish brown, blotched 
with dark between antennal sockets; ocellar spots and clypeus pale; 
palpi pale brown; antenna brown, pale reddish at base; coxae blackish 
brown over most of areas receiving femora in repose; external lateral 
flanges of coxae noticeably paler; remainder of legs uniformly dark 
brown except arolia and bases of claws, which are pale. Supraanal 
plate uniformly dark brown; subgenital plate brown, apical flap pale; 
cercus brown, grading to pale in apical fifth. 

Measurements. Length of body 15 mm., of pronotum 5.8 mm., of 
tegmen 4.3 mm., of hind tibia 5.8 mm.; width of pronotum 7.5 mm., of 
tegmen 4.7 mm. 

Female (allotype). Differing from male as follows: Ventral margins 
of hind femur each with six spurs; supraanal plate (fig. 37) more pro- 
duced posteriorly than in male; subgenital plate as in female of 
ramsdeni. 

Coloration. Same in general as that of male; abdomen very dark on 
apical segments; cerci darker than in male> with only a trace of pale at 
apices. 



GURNEY: CUBAN BLATTIDAE 31 . 

Measurements. Length of body 16 mm., of pronotum 5.8 mm., of 
tegmen 4.S mm., of hind tibia 5.6 mm.; width of pronotum 7.6 mm., of 
tegmen 4.8 mm. 

There is one female paratype that agrees essentially with the allo- 
type. The apex of the supraanal plate is shallowly and broadly emargi- 
nate. The apical segments of the abdomen are not so dark as in the 
allotype, and the middle region of tergum 7 is blotched with pale 
orange. One male nymph and one female nymph, measuring 13. 5 mm. 
and 12 mm., respectively, in body length, are dark brown with legs 
paler. No wing or tegminal pads are present. 

Type locaUty. Pico Turc^uino (south side), Oriente Province, Cuba. 

Type. Museum of Comparative Zoology. 

Paratype. U. S. National Museum, No. 54828. 

The type, allotype, paratype, and two nymphs were all taken at 
type locality, 3,000-5,000 feet altitude, June 1936, by P. J. Darlington. 
The male nymph is deposited in the U. S. National ]Museum. The 
author is glad to name this species in honor of the veteran American 
entomologist, Nathan Banks, Curator of Entomology in the Museum 
of Comparative Zoology. 

The Genus Symploce Hebard 

Symploce Hebard, Trans. Amer. Ent. Soc, vol. 42, p. 355, 1916 (genotype, 
Ischnoptera capitaia Saussure, by original designation). 

This genus contains many species in the warmer regions throughout 
the world, but the eight species now recognized as American are all 
found in the West Indies. One of these, Jwspes (Perkins), occurs on 
the American mainland and in Hawaii as well, but in America at 
present Symploce is essentially West Indian. The male genitalia 
furnish the most important specific characters, and illustrations of 
these structures, for the eight West Indian species, have been given by 
Hebard (1916b) and Rehn and Hebard (1927). A ninth species is now 
added, this from the seacoast of Oriente Province, Cuba. The only 
previously recorded Cuban species is the genotype, capitata, but it is 
less closely related to the new species than is jamaicana (Rehn). 

Symploce munda, new species 

Fig. 39 

The male of this species differs from jamaicana in the structure of 
terga 1, 9, and 10, and that of the subgenital plate. 



32 bulletin: museum of comparative zoology 

Male (holotype). Of average size for the genus; interocular space 
narrower than distance between antennal sockets (about as 2.45 to 
3.6); segment 4 of maxillary palpus slightly shorter than apical 
segment. Pronotum, tegmina, and legs essentially as in jamaicana; 
ulnar vein of wing with two complete and two incomplete rami. 
Tergum 1 with dense tuft of hairs on posterior half, projecting forward, 
similar to that of jamaicana; sparse, slender hairs arranged in an 
oblique row on each side of middle near base poorly developed, not so 
prominent as in jamaicana. Tergum 9 unspecialized (in jamaicana 
there is a dorsally projecting lip, bearing a comb of sparse, erect hairs). 
Supraanal plate similar in general form to that of jamaicana, but 
strongly bilobed apically, unspecialized basally (the apex in jamaicana 
is entire, broadly rounded, with a prominent raised lip at the base of 
the disk corresponding to that of tergum 9). Subgenital plate (fig. 39) 
with left style heavily sclerotized, somewhat concave in basal two- 
thirds in ventral view, the lateral margin briefly recurved, apical 
third slender, twisted to the left, about five coarse teeth on ventral 
margin; large apical plate on right slightly convex in ventral view, a 
blunt apical tooth curving toward left, its margin minutely serrate; 
right latero-posterior angle of main body of subgenital plate with 
sparse, short, bristly setae {in jamaicana the left style is different, the 
large right plate has an acute hook directed to the left, and there is no 
cluster of bristly spines as in munda). 

Coloration. General color pale yellowish; eyes dark brown; ocellar 
spots pinkish; antenna light brown, paler at base; wide lateral and 
narrow anterior borders of pronotum and marginal field of tegmen 
washed with whitish; legs pale, slightly darker on tibiae, tarsi, spines, 
and spurs; a conspicuous, small, black blotch at base of each coxa; 
abdominal sterna 2-6 with similar blotches at each side, a small black 
dot associated with each blotch laterally; terga 1-7 progressively 
darker toward apex of abdomen, the last very dark brown, all with 
pale margins; terga 8-10 pale yellowish, disk of 10 (supraanal plate) 
yellow; subgenital plate pale except for dark setae and left style; 
cerci pale dorsall}- , all except apical segments divided into dark-brown 
basal half and pale apical half ventrally, apical segments all dark 
ventrally. 

Measurements. Length of body 9 mm., of pronotum 2.4 mm., of 
tegmen 10.5 mm., of hind tibia 3.75 mm.; width of pronotum 3.4 mm. 

Female (allotype). General form as in male; width of interocular 
space considerably narrower than distance between antennal sockets 
(about as 3.45 to 4.6); ulnar vein of wing with three complete and 



GURNEY: CUBAN BLATTIDAE 33 

four incomplete rami; apex of supraanal plate with narrow, V-shaped 
incision; apex of subgenital phate shallowly and very broadly concave. 

Coloration. Differing from male as follows: Terga 4-7 conspicu- 
ously darker, pale posterior margins poorly developed but pale lateral 
borders conspicuous; supraanal plate whitish yellow except a dark- 
brown blotch on disk each side of middle; subgenital plate with wide 
lateral and apical whitish-yellow border and a dark blotch and dot on 
each side as in preceding sterna; cercus uniformly blackish brown 
dorsally, same ventrally except that segments in basal two-thirds are 
somewhat pale in their apical portions. 

Measurements. Length of body 10.5 mm., of pronotum 3.3 mm., of 
tegmen 13 mm., of hind tibia 5.2 mm.; width of pronotum 4.3 mm. 

In addition to the type and allotype just described, there is one male 
paratype. It is smaller than the type and differs otherwise as follows : 
Width of interocular area in proportion to distance between antennal 
sockets as 2.9 to 3.4; ulnar vein of left and right wings with two incom- 
plete and one incomplete rami, and one complete and one incomplete 
rami, respectively. Subgenital plate with left style more prolonged and 
conically tapering at apex, rather than twisted and flattened, without 
teeth; apical hook of large right plate less blunt than in type; group of 
spinelike setae constituting a much larger and denser cluster. 

Measurements. Length of body 7.5 mm., of pronotum 2.3 mm., of 
tegmen 8.5 mm., of hind tibia 3.4 mm.; width of pronotum 2.9 mm. 

Type locality. Coast below Pico Turquino, Oriente Province, Cuba. 

Type. Museum of Comparative Zoology. 

Paratype. U. S. National Museum, No. 54829. 

The type and allotype were taken at the type locality, June 2G-30, 
1936, by P. J. Darlington. The paratype was collected by Dr. Darling- 
ton at Maisi, Oriente Province, July 17, 1936. 



Subfamily BLATTINAE 

The Genus Eurycotis Stal 

Eurycotis St§,l, Bihang K. Svenska Vet. Akad. Handl., Ed. 2, No. 13, p. 13, 
1874 (genotype by monotypy, Polyzosteria rufovittata Brunner). 

Eurycotis is richly represented in Cuba, 16 of the 22 West Indian 
species being known from that island. Of these 22 species, 14 were first 
described by Rehn and Hebard (1927) or have since been discovered, 
indicating that others may yet remain unknown to science. The keys, 
descriptions, and illustrations by Rehn and Hebard, together with 



34 bulletin: museum of comparative zoology 

those of the present paper, should permit the identification of most 
specimens, and a fuller treatment of the genus does not seem timely 
until it is possible to consider the species belonging to the North, Cen- 
tral, and South American faunas. Until that time, the limits of the 
genus must remain uncertain, as indicated by Rehn and Hebard (1. c). 
The generic key of those authors and that of Shelford (1910) suffice for 
identification purposes. The former authors mention 4 apparently 
natural groups into which the West Indian species may be divided. 
Of the 16 Cuban species, all except 5 (lacernata, rhodae, opaca, taurus, 
cribosa) belong to the group of "species of glabrous or shining, almost 
impunctate surface, with tegmina lateral or quadrate, but not attin- 
gent, largely with variegated color pattern." 

It should be noted that Blatta guttata Thunberg, described from St. 
Bartholomew, is possibly a species of Eurycotis (see Rehn and Hebard, 
1. c, p. 159) and that Periplaneta occidcntalis Sauss. has been removed 
from Eurycotis, where it was placed by Shelford (1910, p. 12), to Pel- 
matosUpha (see Rehn and Hebard, 1. c, p. 15). 

Key to the W^est Indian Species of Eurycotis 

1 . Tegmina with inner margins attingent or overlapping 2 

Tegmina with inner margins well separated 6 

2. Entire dorsal surface of body and tegmina pitch black; male 

supraanal plate (fig. 18) with apical angles spiniform, recurved; 
female supraanal plate (fig. 17) transverse, broadly emarginate; 
spiniform postero-lateral production of tergum 7 with a lamella 

on inner side (fig. 2). (Cuba) taurus Rehn and Hebard 

Entire dorsal surface of body and tegmina not pitch black, if light 
areas of pronotum are indistinct (some specimens of opaca) the 
supraanal plate differing in each sex (fig. 14, 16) from above; 
spiniform process of tergum 7 without an inner lamella or (in 
opaca, fig. 5) with this poorly developed 3 

3. Body shining dark brown or black, pale lateral margins (fig. 1) 

extending onto base of tergum 6. (Cuba) .... rhodae, new species 
Body not colored as above 4 

4. Tegmina extending to tergum 1 or slightly beyond, as long as 

wide 5 

Tegmina not reaching tergum 1, much wider than long. (Thorax 
with orange lateral borders, sometimes poorly differentiated.) 
(See figs. 5, 14, 16) (Cuba and Isle of Pines) . . . opaca (Brunner) 

5. Abdomen with a pale dorsal border extending along each lateral 

margin; tegmen subquadrate. (Virgin Islands) .improcera Rehn 



GURNEY: CUBAN BLATTIDAE 35 

Abdomen without pale border along lateral margins; apex of teg- 
men oblique. (Antigua) similis Caudell 

6. Pronotum and dorsal surface of abdomen entirely dark. (In some 

species the hind tibia is expanded laterally about midway of its 

length, and with a conspicuous depression as in fig. 4.) 7 

Neither pronotum nor abdomen entirely dark .10 

7. Tegmen conspicuously yellow 8 

Tegmen entirely dark 9 

8. Hind tibia expanded laterally and with a conspicuous depression 

similar to that of tibialis (fig. 4). (Cuba)^ 

ftaripcnnis Saussure and Zehntner 
Hind tibia not so specialized. (Apical abdominal terga closely and 
heavily punctate; first segment of tarsus with pul villus three- 
fourths the ventral length of segment.) (Cuba) 

cribosa Rehn and Hebard 

9. Male supraanal plate (fig. 10) well produced, though transverse; 

male tegmen (fig. 20) rounded at apex; hind tibia in each sex 
expanded laterally about midway of its length, and with a con- 
spicuous depression (fig. 4). (Hispaniola) tibial is Hebard 

]Male supraanal plate (fig. 12) very decidedly transverse; male 
tegmen (fig. 15) acute at apex; hind tibia of female broad, but 
not specialized as above. (Hind tibia of male specialized.) 
(Jamaica)^ lixa Rehn 

10. Pronotum uniformly pale (except for scattered dark flecks), no 

conspicuous dark pattern 11 

Pronotum not uniformly pale 14 

11. Apical five terga of abdomen entirely dark in color 12 

Apex of abdomen not entirely dark (tegmen subquadrate; supra- 
anal plate transverse in male, bicolored). (Cuba) 

galeoides Rehn and Hebard 

12. Exposed portion of tegmen clearly transverse; tegmina separated 

by less than the width of a tegmen. (Cuba) . . caraibea (Bolivar) 

Exposed portion of tegmen lobiform, not transverse; area between 

tegmina usually wider than the width of a tegmen 13 

13. Abdominal terga 2-5 heavily marked with dark at their bases, the 

dark areas ending abruptly before reaching lateral margins and, 
at their farthest extension laterad, covering about three-fourths 

' E. flavipennis S. & Z. is known only from the original description, and, following Rehn and 
Hebard (1. c, p. 179), the present interpretation of the unsatisfactory description had best be 
retained until further information is available. 

2 E. lixa Rehn is based upon two specimens, male and female, taken at New York City on a 
banana ship from Jamaica. For this reason, Jamaica is considered the native home of liia. 



36 bulletin: museum of comparative zoology 

the length of segment, narrowed each side of median hne. 

(Cuba) scalaris Rehn and Hebard 

Abdominal terga 2-5 with narrow dark basal margins, these often 
indistinct. (Cuba) dimidiata (Bolivar) 

14. Pronotum with a definite dark design on the disk, such as in 

figs. 7-9 16 

Pronotum without definite design on disk such as above, but some- 
times with pale lateral borders 15 

15. A dark species with a striking pale longitudinal band extending 

from head to middle of abdominal tergum 6 on each lateral 
margin, closely resembling rhodae (fig. 1) except that tegmina 
are lateral and widely separated; no transverse bars on ab- 
domen. (Cuba) lacernata Cabrera 

A dark species with lateral borders of pronotum mars orange, 
this coloring not extending onto metanotum or abdomen; terga 
2-7 with narrow basal transverse bars of mikado brown. 
(Hispaniola) histrio Rehn 

16. Head with a transverse, dark interocular bar (fig. 19), not with a 

triangular point extending ventrally to level of antennal bases 

or onto face 17 

Head with a dark interocular marking extending triangularly to 
level of antennal bases (fig. 21) or onto face 20 

17. Dorsal surface of abdomen entirely dark except for lateral macula- 

tions on terga 2-6, no transverse bars; basal segment of hind 
tarsus with pulvillus about two-thirds as long as segment. 

(Bahamas) bahamcnsis Rehn 

Dorsal surface of abdomen marked with transverse bars; basal 
segment of hind tarsus with pulvillus not over one-half as long 
as segment 18 

18. Abdominal tergum 6 and supraanal plate uniformly dark; tegmen 

subquadrate, the nearly straight inner margin angulate with 

apical margin. (Cuba) balteata Cabrera 

Tergum 6 not entirely dark, supraanal plate often reddish brown 
or pale on apical two-thirds ; tegmen not subquadrate, inner and 
apical margins continuous instead of angulate 19 

19. Tegmen narrowly lobiform (fig. 8); pronotum with paired dark 

crescent-shaped marks. (Cuba) fugacis, new species 

Tegmen broadly lobiform, shaped much like tegmen of caitdellana 
(fig. 9) ; pronotal pattern horseshoe shaped, with inner margins 
of dark design irregular. (Cuba) 

ferrum-equinum Rehn and Hebard 



GURNEY: CUBAN BLATTIDAE 37 

20. Abdominal terga 1-5 transversely bordered with dark both 

anteriorly and posteriorly, the dark bars united before reaching 
lateral margins, a large pale submarginal maeulation on each 
lateral border of terga 2-5 (fig. 9); dark posterior border of 
pronotum joined to dark blotches on disk. (Cuba) 

caudellana, new species 

Abdomen not marked as above; dark posterior border of pronotum, 

if developed, not joined to dark blotches on disk 21 

2 1 . Tegmen subquadrate ; dark interocular marking extending ventrally 

to a level with ventral margin of antennal sockets, sides of 
triangular marking irregular. (Cuba) 

torquinensis Rehn and Hebard 
Tegmen lobiform (fig. 7) ; dark interocular marking with apex not 
extending ventrally beyond level of dorsal margins of antennal 
sockets, sides of triangular mai'king smooth. (Cuba) 

famelica, new species 



EuRYCOTis RHODAE, new species 
Fig. 1 

This species is closely related to lacemata, which apparently replaces 
rhodae in western Cuba. The coloration of rhodae is so striking that 
the general appearance distinguishes it from all other known species. 
From laccrnaia it differs in the considerably larger size, in the exposed 
portions of the tegmina being transverse and slightly overlapping in- 
stead of lateral, and shaped much as in caudellana (fig. 9), in the lateral 
margins of the female supraanal plate converging less posteriorly, and 
in the color of the ventral surface of the abdomen, in addition to the 
more minor features noted below. 

Female (holotype). Body smooth, shining; tegmina weakly im- 
punctate. Head in frontal view slightly longer than wide, a little longer 
proportionately than in a female paratype of lacemata: interocular 
space noticeably wider than distance between antennal sockets. 
Anterior margin of pronotum briefly truncate directly above occiput, 
latero-anterior angles broadly rounded and depressed hoodlike about 
eyes. Tegmina subquadrate, exposed portions transverse, overlapping, 
extending posteriorly about one-fourth to one-third length of metano- 
tum. Legs essentially as in lacemata; hind tibia moderately inflated, 
slightly less so than in lacemata, surface smooth; pulvillus of basal 
segment of hind tarsus less than one-fourth as long as segment. Ab- 
domen with latero-posterior angles of terga 2-7 acute, those of tergum 



38 bulletin: museum of comparative zoology 

7 very prolonged, with no inner lamella such as in taurus and opaca 
(figs. 2, 5); abdominal sterna similar to those of lacemata; supraanal 
plate wider at base than long, sides less converging than in lacemata, 
and latero-posterior angles well rounded in contrast to abruptly right- 
angled or slightly acute-angled as in lacemata; apical emargination 
broadly obtuse-angulate. Cerci slender, acuminate (right cercus with 
apex deformed). 

Coloration. Pronotum with disk black, on posterior half an indis- 
tinct V-shaped mark of reddish brown with apex pointing forward 
and sides closely appressed, also a few dots of the same color showing 
through ground color of black; broad lateral borders of yellow as in 
fig. 1; margins dark brown, progressively intense anteriorly along 
lateral margins. Head with interocular area of dark brown, extending 
further posteriorly than in lacemata, reaching well onto occiput; a 
brownish transverse band connecting ventral margins of antennal 
sockets and joined with subquadrate ventral extension of interocular 
marking; region of fronto-clypeal suture pale brown, darker at frontal 
pits, joined to transverse facial band by a narrow longitudinal stripe, a 
single large brown dot on each side of median stripe at ventral margin 
of transverse band. (See Rehn and Hebard, 1927, pi. 12, fig. 7, for 
head coloration of lacemata.) Eyes dark; antennae pale brown. Teg- 
men brownish black, on inner five-eighths tinged with reddish, especially 
on inner posterior region; lateral border yellow, edged laterally with 
brown. Legs generally pale; lineate area at base of each coxa, dorsal 
margin of each femur, especially the ventral margins of hind femur in 
apical half, inner margin of middle tibia, spines, apices of claws, and 
apical portions of tarsal segments varying shades of brown; posterior 
tibia and tarsus dark brown, arolium, pulvilli, and base of claws paler. 
Dorsum of abdomen glossy black; four spots of reddish brown indi- 
cated on metanotum; posterior margin of terga reddish brown in re- 
flected light; apical portion of supraanal plate and cerci reddish brown, 
apex of cercus pale; a broad yellow lateral border on terga 2-5 and at 
base of tergum 6, this pale border having a somewhat narrower counter- 
part on ventral surface of abdomen, which otherwise is uniformly 
black. (The venter of the female paratype of lacemata is light reddish 
brown, with a pale border.) 

Measurements. Length of body 29 mm., of pronotum 7.8 mm., of 
tegmen 4.5 mm., of hind tibia 11.5 mm.; width of pronotum 11.5 ram., 
of tegmen 7.5 mm. 

Type locality. Pico Turcjuino (south side), Oriente Province, Cuba. 

Type. Museum of Comparative Zoology. 



GURNEY: CUBAN BLATTIDAE 39 

The only known specimen, the female type, was collected by P. J. 
Darlington in June 1936, at an altitude of 3,000-5,000 feet. This at- 
tractive species is named in honor of ^Nlrs. Rhoda Frank Mislove, of 
the U. S. National INIuseum, who has assisted the writer in the prepara- 
tion of many Orthoptera for study. 

EuRYCOTis TORQUiNENSis Rehn and Hebard 
Figs. 6, 11 

Eurycotis torquinensis Rehn and Hebard, Bull. Anier. Mus. Nat. Hist., vol. 54, 
art. 1, pp. 163-165, pi. 12, figs. 9-12. 

Material here recorded. 1 male, 8 females, 3 nymphal females, Pico 
Turquino (south side), Oriente Province, 3,000-5,000 feet, June 1936; 
1 female, 2 nvmphal females, same data except altitude, 5,000-6,000 
feet (P. J. DarUngton) (M. C. Z. and U. S. N. M.). 

The above male has been compared with the only previously known 
specimen, the male holotype, and found to agree except in features 
ascribable to individual variation. In two specimens the pronotal 
marking agrees with that of the holotype, but the two dark lanceolate 
blotches of the disk are connected anteriorly by a short transverse bar 
in the other specimens. The middle area of tergum 7 has a pale-brown 
blotch in four specimens. In the female taken at 5,000-6,000 feet the 
dark area of the face is connected v/ith the fronto-clypeal bar; the 
head coloration of the other specimens is about as figured by Rehn and 
Hebard (1. c, pi. 12, fig. 10). The venter of the abdomen is variable 
in color. The sterna of the male are dark brown except for pale lateral 
borders on all except the subgenital plate and penultimate sternum, 
the pale borders extending a short way along the posterior margins. 
The venter of the female varies from entirely pale except for the sub- 
genital plate and penultimate sternum, which are brown, to a condi- 
tion in which all sterna are considerably marked with dark brown. 
The basal sterna may be pale, with a small dark blotch on each lateral 
border. The penultimate sternum of most specimens is chestnut 
brown, marked with adjoining irregular, blackish-brown and pale 
bars on each lateral border. Rehn and Hebard (1. c, p. 170) have 
described similar variation in the coloration of the venter of ferrum- 
equinum. Females agree ^ath the male in the essential details of 
tegmina, interocular width, legs, and other features. The female sub- 
genital plate is typical of this section of the genus, and the supraanal 
plate (fig. 16) is as illustrated. The specimen from the 5,000-6,000 
foot level is smaller than the others. Measurements in millimeters of 



40 bulletin: museum of comparative zoology 

this and those of an average female from the 3,000-5,000 foot level are 
as follows: Length of body 18.5, 22; of pronotum 6.1, 6.9; of tegmen 
2.4, 3.3; of hind tibia 8, 9; width of pronotum 8.4, 10. The nymphs 
range from 12 to 19 mm. in length. A pale border surrounds the 
pronotum anteriorly and laterally and extends back to the apex of 
tergum 5. The hind tibiae and tarsi are dark brown. 

EuRYCOTis FAMELICA, new species 
Figs. 7, 21 

This new species is a close relative of haltcata but differs from that 
species in several distinctive color features and in slight, but none the 
less evident, structural features in tegminal shape. 

Femalr (holotype). Size medium for genus; body very smooth and 
shiny. Head about as wide as long in frontal view (fig. 21). Hind 
tibia noticeably swollen, not flattened as in ferrum-equinum but agree- 
ing essentially with balteata (see Rehn and Hebard, 1. c, p. 166, pi. 12, 
fig. 14) except that it is more narrowed near base than there figured. 
Front and middle tibiae slightly swollen. Pulvillus of basal segment of 
hind tarsus one-fifth the ventral length of segment. Nota and ab- 
dominal segments typical of this section of the genus. Tegmen 
obtuse-angulate at junction of inner and apical margins. Supraanal 
plate about as described in caudellana. Subgenital plate and cerci as 
in that species. 

Coloration. Head with irregularly margined, triangular, interocular 
marking of dark Ijrown as illustrated (fig. 21) ; face and clypeus weakly 
marked with pale brown. Pronotal markings brownish black, the 
margins reddish brown. Legs nearly uniformly pale except for brown 
spines; middle and hind coxae each with a moderately weak, oblique 
brown line near base; dorsal margins of middle and hind femora, apex 
and base of hind tibia, and hind tarsus washed with pale brown. 
Tegmen with inner margin bordered with brown. Markings of terga 
1-9 blackish tinged with reddish, especially along margins. Supra- 
anal plate reddish brown, darker at base; cerci brown. Abdominal 
sterna brown, a wide lateral pale border on segments 2-6. 

Measurements. Length of body 23 mm., of pronotum 6.5 mm., of 
tegmen 3 mm., of hind tibia 9.5 mm.; width of pronotum 10.3 mm. 

Type locality. Loma del Gato, Cobre Range, Oriente Province, Cuba. 

Type. Museum of Comparative Zoology. 

The type female was taken July 3-7, 1936, by P. J. Darlington at an 
altitude of about 3,000 feet. In addition to the type, there is a nymphal 



GURNEY: CUBAN BLATTIDAE 41 

male 12 mm. long with the same data. The dorsal surface of the body is 
dark browTi, a continuous pale border including the cerci. The posterior 
half of the pronotal disk and middle areas of the meso- and metanotum 
are of a lighter shade of brown than the general ground color. 

EuRYCOTis FUGACis, new species 
Figs. 3, 8, 13, 19 

This species is related to haltcata, from which it differs in the shape 
of the tegmina and the inflation of the hind tibiae in both sexes, and in 
the shape of the male supraanal plate. The male genitalia of baUeata 
have not been compared with those oifugacis, but differences probably 
occur. The most conspicuous difference in coloration is found on the 
pronotum. In fugacis (fig. 8) there are two curved marks roughly 
resembling parentheses, while haltcata (see Rehn and Hebard, 1. c, 
pi. 13, fig. 1) has two broad lanceolate blotches. Tergum 6 of fugacis 
is not entirely dark, the tegmen has a border of brown along the inner 
margin, the anterior margin of the metanotum is pale, and the venter 
of the abdomen is pale except for the subgenital plate and the middle 
portions of the four sterna preceding it (sternum 8 all dark in male). 
In contrast, tergum 6 of baltcata is entirely dark, the tegmen is merely 
margined with brown, the anterior margin of the metanotum has a 
large dark spot at each side near the meso-posterior portion of the 
corresponding tegmen, and the venter of the abdomen is solidly dark 
except that the five basal sterna have lateral yellow spots. 

Male (holotype). General form as in female (fig. 3); head relatively 
broad, about as wide as long (fig. 19). Tegmen narrowly lobiform as in 
female. Hind tibia flattened, similar to that of ferrum-equinum. 
Pulvillus of basal segment of hind tarsus nearly one-half ventral length 
of segment. Supraanal plate (abnormal, owing to injury) (fig. 13) 
transverse, broadly bilobed apically. Subgenital plate (also abnormal) 
transverse, apparently typical for this section of the genus. Left 
phallomere of concealed genitalia similar to that of torquinensis (Rehn 
and Hebard, 1. c, pi. 12, fig. 11), flattened, slender, recurved at apex; 
no such hook present as in torquinensis just dorsad of subgenital plate 
on right side (1. c, pi. 12, fig. 12), but that sclerite simple; a compli- 
cated right phallomere (fig. 3) in form of a lamellate triangular hook 
with an associated, rather convex, recurved plate at base, this phallo- 
mere differing from the corresponding structure in torquinensis (fig. 6). 

Coloration. Head with interocular bar of blackish brown (fig. 19); 
region between antennal sockets and along fronto-clypeal suture tinged 



42 bulletin: museum of comparative zoology 

with pale brown; eyes black. Dorsal body and tegminal markings of 
blackish brown on yellow ground color as in female (fig. 8) except as 
noted ; each parenthesis-shaped mark on disk of pronotum interrupted 
with pale brown one-third distance from anterior extremity; dark pos- 
terior borders of mesonotum, metanotum, and abdominal terga rather 
more intense in coloration; tergum 7 with narrow pale anterior border. 
Hind tibia brown, paler near spurs; hind tarsus washed with brown, 
darkest near apices of segments; each coxa marked with brownish 
black near base; margins of hind femur and spurs brown; the other leg 
structures pale. Subgenital plate and penultimate sternum brown; 
sterna 6 and 7 with middle area brown, with pale lateral borders; other 
sterna pale except for brown posterior and lateral margins. Supraanal 
plate blackish brown, blotched transversely across middle with reddish 
brown. Cercus dark brown above, more reddish in basal half, and with 
a subapical pale spot, ventral surface similar. 

Measurements. Length of body 21 mm., of pronotum 6.5 mm., of 
tegmen 2.8 mm., of hind tibia 8 mm.; width of pronotum 9.8 mm. 

Female (allotype). Agreeing with male in all important features ex- 
cept as noted. Interocular area slightly wider than in male. Supraanal 
plate as illustrated (fig. 8), apical margin (slightly asymmetrical as 
result of injury) sharply emarginate. Subgenital plate like those of 
related species, obtuse-angulate in lateral view. 

Coloration. General color of body markings reddish brown. Paren- 
thesis-shaped marks on pronotum not interrupted. Tergum 7 pale 
anteriorly only near lateral borders (probably with continuous an- 
terior border when segments are more fully extended). Hind tibia with 
pale longitudinal stripe occupying lateral half of each side next to 
external margin, remainder pale brown. Penultimate sternum pale 
anteriorly on lateral border. 

Measurements. Length of body 23 mm., of pronotum 6.7 mm., of 
tegmen 3 mm., of hind tibia 8.5 mm.; width of pronotum 10 mm. 

Type locality. Buenos Aires, Trinidad Mts., Santa Clara Province, 
Cuba. 

Tyj^e. Museum of Comparative Zoology, 

The type and allotype were taken at an altitude of 2,500-3,500 feet, 
at the type locality. May 8-14, 1936, by P. J. Darlington. 

EuRYCOTis FERRUM-EQUiNUM Relui and Hebard 

Eurycotis ferru7n-eqidrium Rehn and Hebard, Bull. Amer. Mus. Nat. Hist., 
vol. 54, art. 1, pp. 168-171, pi. 13, figs. 2-5. 
Material here recorded. 1 female, 1 nymphal female, mountains 



GURNEY: CUBAN BLATTIDAE 43 

north of Imias, Oriente Province, 3,000-4,000 feet, July 25-28, 1936 
(P. J. Darlington) (M. C. Z.). 

Identification of the somewhat teneral adult has been confirmed by 
comparison with the types. The nymph, 10 mm. long, is pale at the 
apex of the abdomen, and the terga are obliquely pale bordered along 
the lateral margins. The thorax is bordered in a manner typical of the 
nymphs of this section of Eurycotis. 

EuRYCOTis CAUDELLANA, new species 
Fig. 9 

The color pattern of caudellana is very distinctive, and the species 
should be identifiable by the foregoing key without difficulty ; no other 
species closely resembles it. Eurycotis scalaris of eastern Cuba is sug- 
gestive of caudellana in the markings of the abdomen, but differs in 
other color features and in the shape of the tegmina. 

Female (holotj'pe). Medium to small for the genus; smooth and 
shining. Head in frontal view rather sharply triangular; interocular 
area but little wdder than distance betw^een antennal sockets. Hind 
tibia strongly inflated; pronotum and other legs typical of related 
species; pulvillus of basal segment of hind tarsus occupying no more 
than apical fourth of segment. Tegmen lateral, broadly rounded inner 
and apical margins without sign of angulation. Latero-posterior angles 
of abdominal terga about as in rhodae; supraanal plate with sides 
rather straight in apical two-thirds and converging strongly; apical 
angles slightly acute; apical emargination evenly and deeply rounded. 
Abdominal sterna typical of related species; subgenital plate obtuse- 
angulate in lateral view; cercus moderately broad, acuminate. 

Coloration. Head generally pale; eyes black; interocular area dark 
brow^n, posterior margin irregularly excavate, ventral margin roughly 
triangular, extending to level of an imaginary line connecting middle of 
antennal sockets; slightly darker on clypeus and labrum; antenna light 
brown, pale at base. Pronotum marked with black as illustrated in 
fig. 9; two pale-brown dots on posterior half of disk; a wide lateral 
border of yellow; dark-brown lateral and apical margins. Meso- and 
metanotum and tegmina marked with dark reddish brown as illus- 
trated. Legs generally pale; middle and hind femora each with oblique 
bar of dark brown basally; margins of middle and hind femora, all 
spines, apical and basal portions of hind tibia, and most of hind tarsus 
except pulvilli light brown. Terga yellow, marked with black tinged 
with reddish as illustrated; lateral margins brown; supraanal plate 
brownish toward apex; cercus reddish brown, darker along margins; 



44 bulletin: museum of comparative zoology 

sterna brown, the posterior segments with pale lateral borders; sub- 
genital plate somewhat darker at apex. 

Measurements. Length of body (abdomen loosely attached) 21 mm., 
of pronotum 6 mm., of tegmen 2.4 mm., of hind tibia 12 mm.; width of 
pronotum 13.2 mm. 

Type localitij. Rio de Auras (about 65 miles southwest of Cardenas), 
near Union de Reyes, Matanzas Province, Cuba. 

Type. U. S. National Museum, No. 54328. 

The female holotype is the only known specimen. 

Probably no entomologist more conscientiously devoted a large part 
of his life to a study of the literature of his favorite group of insects 
than did the late Andrew Nelson Caudell during the 35 years that 
Orthoptera and the literature concerning them were both his joy and 
his life work. Not privileged to know him personally, the writer appre- 
ciates Mr. Caudell's contributions, especially through the good fortune 
of having daily contact with the literature reference catalogue pre- 
pared by him. The present species is named caudellana in respect for 
his memory. 

EuRYCOTis CARAIBEA (Bohvar) 

Polyzosteria caraibea Bolivar, Mem. Soc. Zool. France, vol. 1, p. 126, 1888. 

Material here recorded. 1 male. Upper Ovando River, Oriente Prov- 
ince, 1,000-2,000 feet, July 17-20, 1936; 1 male, mountains north of 
Imias, Oriente Province, 3,000-4,000 feet, July 25-28, 1936 (P. J. 
Darlington); 1 male, introduced at Brainerd, Minn., July 1921. 
(M. C. Z. and U. S. N. M.) 

Rehn and Hebard (1927, p. 178, pi. 14, figs. 3, 4) have given descrip- 
tive notes on this species. In the original description, based on a male, 
Bolivar gave the following length measurements: Body 21 mm., 
pronotum 7.5 mm., tegmen 3 mm. In the above three specimens the 
pronotum is 7.5 to 8 mm. long and the body length is 24, 24, and 30 
mm., respectively, owing to different degrees of body extension. The 
Monte Libano male recorded by Rehn and Hebard has been examined ; 
the genitalia agree with those of the specimens here recorded. The 
phallomeres include several genital hooks none of which resembles the 
right hooks oi fugacis and torquincnsis (figs. 3, 6). 

EuRYCOTis TAURUS Rchn and Hebard 

Figs. 2, 17, 18 

Eurycoti.^ tanrus Rehn and Hebard, Bull. Anier. Mus. Nat. Hist., vol. 54. art. 1, 
pp. 182-185, pi. 14, figs. 5-8, 1927. 



gurnet: CUBAN BLATTIDAE 45 

Material here recorded. 1 female, Coast below Pico Titrquino, Oriente 
Province, June 26-30, 1936 (P. J. Darlington) (M. C. Z.). 

Because of its size and pitch-black color, in addition to the distinctive 
features illustrated, this large species is not likely to be confused with 
any other. In the above-mentioned specimen there is a small, con- 
spicuous spot of orange at each latero-anterior angle of the subgenital 
plate. Other features agree with the original description. Comparison 
with the type, allotype, and two paratypes has been made. 

Subfamily EPILAMPRINAE 

The Genus Audreia Shelford 

Audreia Shelford, Gen. Insect., Fasc. 101, p. 11, 1910 (genotype, Calolampra 
carinulata Saussure, designated by Hebard, 1920, p. 92). 

In addition to the two Cuban species separated in the following key, 
Audreia contains j a maicana R. & H., several species from the mainland 
of tropical America, and several Old World species that have been 
referred to this genus. 

Key to the Cuban Species of Audreia 

Tegmina attingent or slightly overlapping; each tegmen subquadrate, 
posterior margin oblique and usually concave; posterior margin of 
pronotum very broadly and obtusely angulate in middle 

hamiltoni (Rehn) 

Tegmina lateral; each tegmen (figs. 41, 43) lobate; posterior margin 
of pronotum truncate exploratrix, new species 

Audreia il^miltoni (Rehn) 

Cahlampra hamiltoni Rehn, Trans. Amer. Ent. Soc, vol. 29, pp. 274-275, 1903. 

. Material here recorded. 1 male, 6 females, 4 nymphs, Pico Turquino 
(south side), Oriente Province, 3,000-5,000 feet, June 1936; 1 nymph, 
same, 5,000-6,000 feet, June 1936 (P. J. Darlington) (M. C. Z. and 
U. S. N. M.). 

The above male agrees in general with that described by Rehn and 
Hebard (1927, p. 205, pi. 15), but the pronotal disk is shining black 
and no lyrate marking is evident. Projecting near the right style is a 
phallomere, long, flattened, hooked like a shepherd crook apically, the 
very tip curved laterally; about 20 minute lateral bristles basad of 
hooked portion. 



46 bulletin: museum of comparative zoology 

Measurements. Length of body 16.5 mm., of pronotum 5.5 mm., of 
tegmen 5 mm., of hind tibia 6.7 mm.; width of pronotum 8 mm., of 
tegmen 5.5 mm. 

The posterior margin of the pronotum varies from slightly to 
markedly concave in the females. The pronotal disk is very dark in all 
but one, which has the lyrate pattern well developed. Body length 
ranges from 19.5 to 23 mm. The nymphs range from 9 to 17 mm. in 
body length, and the pronotum has the dark disk and paler lateral 
borders characteristic of the species. 

Audreia exploratrix, new species 
Figs. 41-43 

This is a close relative of hamiltoni, differing from that species in the 
color of the interocular and interocellar areas, in addition to the 
features mentioned in the key. 

Male (holotype). Interocular area wider than distance between 
antennal sockets (as 4.7 to 4), slightly wider proportionately than in 
male of hamiltoni. Segments of maxillary palpus short; apical segment 
slightly less than twice as long as greatest width, length about equal 
to segment 3, slightly less than segment 4. Pronotum in cross section 
vaulted and domelike, somewhat as in hamiltoni, dorsal outline as in 
fig. 43; a shallow, transverse depression, about as long as distance be- 
tween tegmina, on disk just in front of posterior margin of pronotum; 
the whole sparsely punctate. Tegmen as in fig. 43 in dorsal view, in 
dorso-lateral view inner-apical margin nearly straight-oblique, costal 
margin gently convex, venation about as in female paratype (fig. 41); 
wing vestigial, reaching nearly to apex of tegmen. Legs as in hamiltoni. 
Terga 2-7 with latero-posterior angles not spined, but rounded or sub- 
rectangular; longitudinal weltlike ridges present. Supraanal plate 
bilobed, moderately emarginate. Subgenital plate (fig. 42) asym- 
metrical, the styles slender and delicate, the left style not located so 
near apex of plate as in hamiltoni but this difference perhaps due to 
individual variation. 

Coloration. General color dark brown, with areas of lighter color 
and mottling typical of this section of Audreia. Head in frontal view 
black from postclypeus to an imaginary transverse line bisecting inter- 
ocular area (in hamilto7ii the black color of face extending dorsally 
only slightly onto interocular area); occiput with light mottling, 
blotched with dark brown at base; eyes grayish black; cheeks pale 
brown; labrum and anteclypeus yellow; each ocellar spot and narrow 



GURNEY: CUBAN BLATTIDAE , 47 

stripe extending to eye and thence along inner margin of eye to occiput 
pale; maxillary palpus brown, darker on segment 5; antenna brown, 
segment 2 paler. Pronotum with disk blackish brown, faintly tinged 
with red; a trace of lyrate marking in a dark longitudinal median 
mark on anterior half of disk; borders of yellowish orange, heavily 
spotted with dark brown. 

Tegmen dark brown on inner two-thirds of basal half, reddish orange 
on narrow inner border and on inner portion of apical half, border along 
costal margin yellowish; the light areas with small dark blotches. Each 
coxa pale, dark brown at base, the surface receiving femur in repose 
darkly punctate; front femur nearly all brown, faces paler; middle and 
hind femora yellow, external margins, indefinite longitudinal facial 
stripes, and sockets of spurs on inner margins dark brown ; front tibia 
mostly brown, blotched with pale near apex; middle and hind tibiae 
with brown and pale longitudinal streaks, darkest at bases of spurs; 
front tarsus mostly brown, pulvilli and bases of segments 1 and 5 pale; 
middle and hind tarsi mostly pale, dark in apical portions of segments, 
especially on lateral surfaces. 

Dorsal surface of abdomen mottled much as in hamiltoni, but some- 
what darker, pale areas more conspicuous toward apex; basal half of 
supraanal plate dark brown, a pale area in middle at base, apical half 
pale brown, sparse dark punctations; venter of abdomen uniformly 
very dark, shining brownish black; subgenital plate narrowly margined 
with light brown apically, styles pale. Dorsal surface of cercus brown- 
ish, pale at base; ventral surface brownish, apex darker. 

Measurements. Length of body 18 mm., of pronotum 5 mm., of 
tegmen 3.6 mm., of hind tibia 5 mm.; width of pronotum 7.1 mm. 

Female (allotype). Differing essentially from male as noted. Ratio 
of interocular area to distance between antennal sockets 5.6 to 4.5; 
pronotum with a broad, shallow depression on posterior half of disk; 
costal margin of tegmen less convex, only slightly convex in dorso- 
lateral view; terga 2-7 with latero-posterior angles acute to spined, 
spines best developed on 5-7; supraanal plate more broadly rounded, 
weakly emarginate apically; subgenital plate evenly rounded apically, 
each lateral margin broadly concavo-emarginate opposite cercus as in 
haviihoni. 

Coloration. Much as in male, but considerably darker on dorsal 
surface of pronotum and tegmina; disk of pronotum with lighter bor- 
ders reduced in size; tegmen blackish brown except basal three-fourths 
of costal margin, which is yellowish orange with sparse dark spots; 
tibiae and tarsi somewhat darker than in male ; coxae, femora, and dor- 



48 bulletin: museum of comparative zoology 

sum of abdomen about as in male; supraanal plate with a dark-brown 
elongate blotch in middle of basal half; lateral portions of basal half 
pale, with dark spots; apical half darker; subgenital plate uniformly 
dark; cercus dark brown, dorsal surface pale at base. 

Measurements. Length of body IS mm., of pronotum 5.7 mm., of 
tegmen 4.5 mm., of hind tibia 5.5 mm.; width of pronotum 8.5 mm. 

There is one female paratype which agrees essentially with the allo- 
type. The costal margin of the tegmen (fig. 41) is barely concave when 
seen in dorso-lateral view. The tegmen is colored about as in the male, 
the dorsum of the abdomen is rather darker than in the allotype, and 
the supraanal plate is all dark brown except for a few weakly developed 
orange spots at the base. The abdominal segments are more extended 
than in the allotype. Measurements. Length of body 23 mm., of pro- 
notum 6.5 mm., of tegmen 5 mm., of hind tibia 7 mm.; width of pro- 
notum 9 mm. 

Type locality. Buenos Aires, Trinidad Mts., Santa Clara Province, 
Cuba. 

Type. Museum of Comparative Zoology. 

Paratype. U. S. National Museum, No. 54830. 

The type, allotype, and paratype were taken at the type locality, 
2,500-3,500 feet. May 8-14, 1936, by P. J. Darlington. 

The Genus Epilampra Burmeister 

Epilampra Burmeister, Handb. Ent., Bd. 2, Abt. 2 (1st halfj, p. 504, 1838 
(genotype, Blatta brasiliensis Fabricius, designated by Kirby, 1903, p. 276). 

There is considerable confusion in the literature concerning the geno- 
type of Epilampra and its identity. Kirby's designation of brasiliensis 
appeared in August 1903, and in the same paper he designated Epi- 
lampra nebulosa Burm. as type of a new genus, Pseudophoraspis. In 
the following month Rehn (1903, p. 271) designated nebulosa as type of 
Epilampra. Kirby's action has priority. In his 1904 catalogue, how- 
ever, Kirby gave Blatta maculicollis Serv. as type of Epilampra, placing 
brasiliensis as treated by Brunner (1865) and Burmeister (1838) in 
synonymy (the latter with a query), under the belief that the species 
called brasiliensis F. by Burmeister was misidentified. Shelford (1906, 
p. 276) has compared drawings in the Hope Museum at Oxford, pre- 
pared from Serville's specimens of maculicollis by Westwood, with 
Fabricius' type of brasiliensis in the Banksian cabinet of the British 
Museum, and he considers the two species distinct. Regardless of mis- 
identification by Burmeister, which remains uncertain, the actual 



GURNEY: CUBAN BLATTIDAE 49 

brasilie7isis of Fabricius is the genotype of Epilampra by Kirby's 
1903 designation. 

As pointed out by Hebard (1920) and Rehn and Hebard (1927), the 
limits of Epilampra and Audrcia are poorly known and need thorough 
re\'ision. Until that is done, the generic placement of E. cuhensis Bol. 
is uncertain, and for the present this species is left in its original genus. 

The present collection contains four of the five known Cuban species 
of Epilampra. Because of the comprehensive manner in which the 
group has been dealt with by Rehn and Hebard no key is presented 
here. 

Epilampra cubensis Bolivar 
Fig. 40 
Epilampra ciibcttsis Bolivar, Mem. Soc. Zool. France, vol. 1, p. 127, 1888. 

Material here recorded. 1 female, Cuchillo de Guajimero (near Imias), 
Oriente Province, about 2,000 feet, July 25, 1936 (P. J. Darlington) 
(M. C. Z.). 

This specimen is somewhat longer than indicated in Bolivar's 
description but otherwise agrees essentially. The general form is 
robust and suggestive of species of Pelmatosilpha though definitely 
epilamprine in its characters. 

Gundlach (1890-91, p. 308) says that material of cubensis (very 
likely Bolivar's type material) came from Mata (about 10 miles south- 
east of Baracoa, Oriente Province). 

Female. Head in frontal view as broad as long; width of interocular 
space and distance between antennal sockets subequal; face well filled, 
convex, with a weak, irregular, transverse depression between ventral 
margins of ocellar spots; apical segment of maxillary palpus with 
ventral margin broadly convex, length compared to segments 4 and 3 
as 5 to 4 to 4; antenna normal for genus. 

Pronotum not quite concealing occiput in dorsal view; point of 
greatest width near latero-posterior angles; posterior margin broadly 
and very obtusel}^ angulate, barely concave on each side between 
middle and a point opposite anal sulcus of tegmen; dorsal surface 
smooth, moderately shining; disk with small depression on each side of 
middle corresponding to part of obscure lyrate pattern; front margin 
very narrowly marginate, gradually more widely so to latero-posterior 
angles, narrowed near base of costal margin of tegmen, then gradually 
widening again to middle of posterior margin. Tegmen (fig. 40) with 
venation as illustrated, marginate along costal margin, reaching about 



50 bulletin: museum of comparative zoology 

to tergum 4. Wing poorly developed, reaching onto basal part of 
tergum 3, entirely covered by tegmen ; venation well formed but wing 
probably non-functional, about 9 rami of ulnar vein visible. Legs 
typically epilamprine ; middle region of front ventral margin of front 
femur with 4 subequal sturdy spines followed by about 12 tiny setalike 
spines and 2 genicular spurs ; basal segment of hind tarsus longer than 
remaining segments; tarsal claws unspecialized; arolium of moderate 
size. Dorsal surface of abdomen flattened, shining; longitudinal welt- 
like ridges poorly developed but evident; latero-posterior angles of 
tergum 2 about right-angled ; angles of terga 3-7 more acute, becoming 
sharply spiniform; supraanal plate deeply incised, broadly bilobed; 
cercus fairly slender, gradually tapering in apical half, acute at tip; 
sub-genital plate unspecialized, evenly rounded apically, each side 
broadly concave near base of cercus. 

Coloration. Interocular space blackish brown, grading to dark 
chestnut on posterior part of occiput; blackish brown on face and 
postclypeus; a small chestnut blotch on front between ocellar spots; 
anteclypeus, labrum, most of mandibles, ocellar spots, margins of 
eyes, and area connecting each ocellar spot with corresponding eye 
pale; cheeks fuscous; eyes grayish black; maxillary palpus light brown, 
apical segment darker. Pronotum chestnut, slightly darker on pos- 
terior border; a few dark-brown spots on disk representing lyrate 
pattern. Tegmen chestnut; scapular and marginal fields lighter, 
mottled with rust-colored spots; a short black line in depression 
laterad of base of anal sulcus ("vena anali linea nigra apposita" — 
Bolivar). 

Front coxa mostly pale, darker at base and on dorsal portion of sur- 
face receiving femur in repose; middle and hind coxae each averaging 
darker, punctate with black on surface receiving femur; front leg be- 
yond coxa mostly fuscous, with pale streaks; middle and hind 
femora mostly yellowish, streaked with fuscous ; middle tibia streaked 
with fuscous and yellowish, darker at bases of spurs; hind tibia very 
dark brown, spurs lighter; middle and hind tarsi mostly light fuscous, 
the basal segment of hind tarsus especially dark. Dorsum of abdomen 
brownish black, tinged with chestnut along lateral margins and on 
basal half of supraanal plate. Cercus pale, darker at apex. Venter of 
abdomen brownish black, the segments slightly margined laterally 
with chestnut and (in reflected light) with traces of paler posterior 
margins; subgenital plate dark, with lateral margins, a weak apical 
margin, and a small spot each side of middle on apical half light 
chestnut. 



gurnet: CUBAN BLATTIDAE 51 

Measurements. Length of body 23 mm., of pronotum 7.4 mm., of 
tegmen 11 mm., of hind tibia 8.5 mm.; width of pronotum 9.6 mm., 
of tegmen 6.7 mm. 

Epilampra tainana Rehn and Hebard 

Epilampra tainana Rehn and Hebard, Bull. Anier. Mus. Nat. Hist., vol. 54, 
art. 1, pp. 213-216, pi. 16, figs. 9-11, 1927. 

Material here recorded. 1 male, mountains north of Imias, Oriente 
Province, 3,000-4,000 feet, July 25-28, 1936 (P. J. Darlington) 
(M. C. Z.). 

The male has previously been unknown, and the dorsal surface of 
this specimen is of a lighter color than the type series, and the abdomen 
is darker beneath, but agrees essentially in facial pattern and other 
features. 

Male. Interocular area subequal to distance between ocellar spots 
and considerably narrower than distance between antennal sockets 
(as 2.4 to 3.4); apical segment of maxillary palpus three times as long 
as wide, noticeably longer than segment 4; supraanal plate extending 
well beyond subgenital plate, rather deeply incised and prominently 
bilobed apically; subgenital plate asymmetrical, a shallow concavity 
on apical margin to the left of middle, margin sinuate on right side 
posterior to right style. 

Measurements. Length of body 18 mm., of pronotum 5.5 mm., of 
tegmen 20 mm., of hind tibia 7 mm. ; width of pronotum 6.8 mm. 

Epilampra gundlachi Rehn and Hebard 

Epilampra gundlachi Rehn and Hebard, Bull. Amer. Mus. Nat. Hist., vol. 54, 
art. 1, pp. 223-225, pi. 17, figs. 4 and 5, pi. 18, figs. 1 and 2, 1927. 

Material here recorded. 1 male, 2 females, mountains north of Imias, 
Oriente Province, 3,000-4,000 feet, July 25-28, 1936; 1 female. Gran 
Piedra Range (north of Daiquiri), Oriente Province, 2,000-3,000 feet, 
May 30-31, 1936 (P. J. Darlington) (M. C. Z. and U. S. N. M.). 

The above male differs from the type, with which it has been com- 
pared, in having the tegmina longer, the pronotum somewhat darker, 
and the dark blotch on the subgenital plate smaller. 

Measurements. Length of body 18 mm., of pronotum 4.7 mm., of 
tegmen 19.5 mm., of hind tibia 6.8 mm.; width of pronotum 6 mm. 

x^mong the three females there is little variation. The interocular 
width is slightly greater than interocellar width and considerably less 



52 bulletin: museum of comparative zoology 

than the distance between the antennal sockets. The Gran Piedra 
specimen has a dark transverse bar on the ventral part of the face, 
with two dark extensions, one at each end of the bar, onto the post- 
clypeus. The venter of the abdomen in these specimens is sprinkled 
with fuscous. As in the male, there is a small premarginal dark blotch 
on each side near the base of each segment. The subgenital plate is 
similar to that of related species and, in addition to a sprinkling of 
fuscous, has four small, dark blotches on the basal part of the disk. 
The supraanal plate is bilobed apically, with a transverse dark mark 
subbasally. 

Measurements of the Gran Piedra female: Length of body 18 mm., 
of pronotum 5.1 mm., of tegmen 22 mm., of hind tibia 7.5 mm.; width 
of pronotum 6.6 mm. 



Epilampra burmeisteri (Guerin) 

Blatta (Phyllodromia) burmeisteri Guerin, in La Sagra, Hist. Phys. Polit. 
Nat. Cuba, pp. 345-346, 1857. 

Material here recorded. 1 female, Upper Ovando River (south of 
Cape Maisi), Oriente Province, 1,000-2,000 feet, July 17-20, 1936; 
1 male, Yunque de Baracoa, Oriente Province, 1,000-1,800 feet, 
July 13, 1936; 1 male, Pico Turquino (south side), 3,000-5,000 feet, 
June 1936 (P. J. Darlington) (M. C. Z. and U. S. N. M.). 

This species is especially distinguished by its small size and con- 
trastingly colored tegmina with pale marginal and scapular fields and 
very dark basal portion of humeral trunk. The postclypeus of two of 
the above specimens is mostly dark. There is slight variation in the 
color of the face and interocular area, but there are no marked differ- 
ences from the head illustrated by Rehn and Hebard (1927, pi. 17, 

fig. 6). 

Measurements of the Turquino specimen. Length of body 14.5 mm., 
of pronotum 4.3 mm., of tegmen 16.5 mm., of hind tibia 5.7 mm.; 
width of pronotum 5.2 mm. The tegminal length of the Yunque de 
Baracoa specimen is 15 mm. and the body length 13.5 mm. 



Epilampra species 

Material here recorded. 1 nymph. Upper Ovando River, Oriente 
Province, 1,000-2,000 feet, July 17-20, 1936; 3 nymphs, mountains 
north of Imias, Oriente Province, 3,000-4,000 feet, July 25-28, 1936; 



GURNEY: CUBAN BLATTIDAE 53 

3 nymphs, Pico Turquino (south side), Oriente Province,. 4,500-6,000 
feet, June 18-20, 1936 (P. J. Darlington) (M. C. Z. and U. S. N. M.). 

These specimens range from 6.5 mm. to 19 mm. in body length and 
cannot be identified to species with certainty. 



Subfamily PANCHLORINAE 

The Genus Pycnoscelus Scudder 

Pijcnoscelus Soudcler, Best. Journ. Nat. Hist., vol. 7, No. 3, p. 421, 1862 (geno- 
type, Pycnoscelus obscurus Scudder (synonym of Blatta surinamensis L.j, 
by monotypy). 

Pycnoscelus surinamensis (Linnaeus) 

Blatta surinanu'Hsis Linnaeus, Syst. Nat., ed. 10, p. 424, 1758. 

Material here recorded. 1 female, Upper Ovando River, Oriente 
Province, 1,000-2,000 feet, July 17-20, 1936; 3 females, coast below 
Pico Turquino, Oriente Province, June 26-30, 1936 (P. J. Darlington); 
1 male, 1 female, Bellevue, Antigua, Sept. 20, 1937 (H. E. Box); 
1 male, Bangkok, Siam, March 1-24, 1932 (Hugh Smith). (M. C. Z. 
and U. S. N. M.). 

It should be noted that this widespread tropical roach was first 
described by Linnaeus in 1758, and not in Edition 12 of the Systema 
Naturae as has been stated several times. It is rather remarkable that 
the genus Pycnoscelus was based upon an adventive immature female 
discovered in the hills of western Massachusetts. The author has not 
seen Scudder's unique type of the synonymous obscurus, but the 
description agrees so well with surinamensis nymphs of the same size 
(being very different from native New England Parcohlatta) that the 
correctness of the synonymy, generally accepted for many years, is 
evident. 

The males of surinamensis are extremely rare and as far as the writer 
is aware were unknown from the New World until Davis (1919, p. 109) 
reported finding a male in the Reptile House of the New York Zoologi- 
cal Society. This may have been introduced from the Old World, 
however. If so, the Antigua specimen recorded above is the first Ameri- 
can male. Hebard (1917, p. 194) has described a male from the Lesser 
Sunda Islands, and the two males in the National Museum agree with 
his description. 

Blatchley (1920, p. 105), apparently not realizing that immature 



54 bulletin: museum of comparative zoology 

female roaches may have styles on the subgenital plate preceding the 
final nymphal instar, says that it "passeth understanding" how the 
specimen of ohscurus described by Scudder could be a female. Sharp 
(1895, p. 224) pointed out that young female roaches may have styles, 
and Quadri (1940, pp. 140-142) shows that female nymphs of Peri- 
planeta americana (L.) have styles until the last instar preceding 
maturity. Other writers have also referred to the matter. Caudell 
(1925) described female nymphs of surinavieiisis and reported rearing 
experiments which showed that this species can and does produce 
young parthenogenetically and that the oothecae are seldom extruded 
from the mother's body, the young usually being born alive. 

This roach is of economic importance as intermediate host for 
Manson's eye worm (Oxi/spirura) of poultry and many wild birds (see 
Shealy, 1927). It is also well known as a greenhouse pest in the north- 
eastern part of the United States. 



Subfamily BLABERINAE 

The Genus Byrsotria Stal 

Byrsotrin St&l, Biliang Till K. Svenska Vet. Akad. Handlingar, Bd. 2, No. 13' 
p. 18, 1874 (genotype, Blatta thimgbergii Guerin (synonym of Blatta 
fvmigala Guerin), by monotypy). 

Byrsotria cabrerai Rehn and Hebard 

Byrsotria cabrerae Rehn and Hebard, Bull. Amer. Mus. Nat. Hist., vol. 54, 
art. 1, pp. 266-268, pi. 24, figs. 1 and 2, 1927. 

Material here recorded. 1 male, 1 female, coast below Pico Turquino, 
Oriente Province, June 20-30, 1936 (P. J. Darlington) (M. C. Z.); 
1 male, between Cardenas and Varadero, Matanzas Province, Novem- 
ber 1931 (H. A. Pilsbry) (A. N. S. P.). 

This species was named in honor of Senor Jose Cabrera, and, because 
of the rules regarding the formation of specific names derived from a 
man's name, it seems best to emend the original spelling, which may 
have been. accidental, to cabrerai. 

The wing and tegminal length are variable ; the tegmina of the Tur- 
quino male only slightly surpass tergum 4, and the wings do not extend 
beyond tergum 2, while the tegmina of the male in the Philadelphia 
collections recorded above are of about the same length. This condi- 
tion contrasts with that of the holotype, which has tegmina reaching 
onto tergum 7 and well developed wings. 



GURNEY: CUBAN BLATTIDAE 55 

Measurements in millimeters of the Turquino male and female, 
respectively, are as follows: Length of body 33, 44.5; of pronotum 10.5, 
13.5; of tegmen 16.5, 12.3; of hind tibia 12, 15; width of pronotum 16.5, 
23. 

The male from Matanzas Province indicates that cabrerai occurs in 
western as well as eastern Cuba. 



Subfamily CORYDIINAE 

The Genus Holocompsa Burmeister 

Holocompsa Burmeister, Handb. Ent., Bd. 2, Abt. 2 (1st Half), p. 491, 1838 
(genotype, designated by Kirby, 1904, p. 169, Conjdia collaris Burm. 
(synonym of Blatta nitidula F.).) 

Holocompsa nitidula (Fabricius) 
Blatta nitidula Fabricius, Species Insectorum, vol. 1, p. 34.5, 1781. 

Material here recorded. 1 female, Loma del Gato, Cobra Range, 
Oriente Province, about 3,000 feet, July 3-7, 1936 (P. J. Darlington) 
(M. C. Z.). 

With the pubescent, orange pronotum, and tegmen divided obliquely 
into bluish metallic and hyaline areas, the present specimen is typical 
of niiidida, the male of which has a blackish pronotum. This domicili- 
ary species apparently occurs even in the higher Cuban mountains. 
The only other described West Indian species of Holocompsa is metallica 
R. & H. of Hispaniola. 

The Genus Pholadoblatta Rehn and Hebard 

Pholadoblatta Rehn and Hebard, Bull. Amer. Mus. Nat. Hist., vol. 54, art. 1, 
p. 286, 1927 (genotj^pe, Aphlcbia inusitata Rehn, by original designation 
and by monotypy). 

Pholadoblatta inusitata (Rehn) 

Fig. 33 

AphleMa inusitata Rehn, Bull. Amer. Nat. Hist., vol. 22, art. 5, p. 113, 1906. 

Material here recorded. 1 male, Soledad' (near Cienfuegos), Santa 
Clara Province, May 1936 (P. J. Darlington) (M. C. Z.). 

1 See Barbour and Robinson (1940) for a general account of the laboratory of Harvard Uni- 
versity located at Soledad. 



56 bulletin: museum of comparative zoology 

This specimen differs from the type (Rehn and Hebard, 1927, pi. 20, 
fig. 6) in having the apical margin of the tegmen broadly and evenly 
rounded, thus agreeing with the Camoa specimen recorded by the 
latter authors. There is a well sclerotized, gently curved phallomere 
dorsad of the right half of the subgenital plate. The left style is direc- 
ted ventro-posteriorly and is more elongate and gradually tapering 
than shown in fig. 33 on account of foreshortening. 



Subfamily OXYHALOINAE 

The Genus Plectoptera Saussure 

Plectoptera Saussure, Mem. Hist. Nat. Mex., Mem. 4, ]). 173, 1864, (genotype, 
designated by Rehn, 1903, p. 281, Blatta porceUana Saussure). 

Of the 10 West Indian species of Plectoptera treated by Rehn and 
Hebard (1927), the following 4 occur in Cuba: ■porceUana (Sauss.), 
vermiculata R. & H., lacerna R. & H., and poeyi (Sauss.). Because the 
differences in color and in male genitalia were freely illustrated by 
Rehn and Hebard (I. c), a key is not presented here. 

Plectoptera porcellana (Saussure) 

Plectoptera porcellana Saussure, Kev. at. Mag. ZooL, Ser. 2, Vol. 14, p. 164' 
1862. 

Material here recorded. 2 males, Loma del Gato, Cobre Range. 
Oriente Province, about 3,000 feet, July 3-7, 1936; 1 male, 1 female, 
Pico Turquino, 5,000-6,000 feet, June 1936; 1 male, same (south side), 
3,000-5,000 feet; 1 female, Buenos Aires, Trinidad Mts., Santa Clara 
Province, 2,500-3,500 feet, May 8-14, 1936 (P. J. Darlington) (M. C. 
Z. and U. S. N. M.). 

In the color of head, pronotum, and tegmina and in the directions in 
which the appendages of the male subgenital plate are bent the varia- 
tion in these specimens agrees with previous information about 
porcellana. It is of interest to find porcellana near the summit of 
Turquino, and it probably occurs throughout Cuba. 

SUMMARY 

This paper is a report on a collection of roaches made in Cuba during 
1930 by P. J. Darlington. Sixteen genera are treated, and an attempt 
has been made to consider each comprehensively as far as Cuba is 



GURNEY: CUBAN BLATTIDAE 57 

concerned; in several cases the entire West Indian fauna or an entire 
genus is reviewed. A new subspecies of Ischnoptera ligula R. & H. is 
described and 9 new species, the latter in the genera Neoblattella, 
Ischnoptera, Nelipophygus, Symploce, Eurycotis, and Audreia. The 
consideration of Eurycotis, including a key to the 22 West Indian 
species, descriptions of 4 new species, and 2 plates of illustrations, is 
one of the more comprehensive sections of the report. Several species 
are discussed about which very little has previously been known. 
Variation has been considered important in the treatment of species, 
and the generic diagnosis of Nelipophygus has been modified so that a 
new species may be referred to that genus. A list of West Indian 
Blattidae described since 1927, when a revision by Rehn and Hebard 
appeared, has been compiled. Because of homonymy, a new specific 
name, tepperana, is given to Blatta rufa (Tepper) of Australia. 



58 bulletin: museum of comparative zoology 



LITERATURE CITED 

Barbour, Thomas 

1923. The birds of Cuba. Mem. Nuttall Ornith. Club, No. 6, 141 
pp., ill. 

Barbour, Thomas and Robinson, Helene M. 

1940. Forty years of Soledad. Sci. Month., vol. 51, pp. 140-146, ill. 

Bates, Marston 

1935. The butterflies of Cuba. Bull. Mus. Comp. ZooL, vol. 78, No. 2, 
pp. 63-258, ill. 

Blatchley, W. S. 

1920. Orthoptera of Northeastern America. IndianapoHs, 784 pp., ill. 

Brunner de Wattenwyl, C. 

1865. Nouveau Systeme des Blattaires. Vienna, 426 pp., ill. 

Burmeister, H. 

1838. Handbuch der Entomologie, Bd. 2, Abt. 2 (1st. Half), 760 pp., 
Berlin. 

Caudell, a. N. 

1925. Pycnoscelus surinamensis Linnaeus (Orthoptera); on its nymphs 
and the damage it does to rose bushes. Proc. Ent. Soc. Wash., 
vol. 27, pp. 154-157, ill. 

Davis, W. T. 

1919. The males of the roach, Pycnoscelus surinamensis. Joum. N. Y. 
Ent. Soc, vol. 27, pp. 108-109. 

Darlington, P. J. 

1937. West Indian Carabidae III: New species and records from Cuba, 
with a brief discussion of the mountain fauna. Mem. Soc. Cubana 
Hist. Nat., vol. 11, pp. 115-136. 

1938. The origin of the fauna of the Greater Antilles, with discussion of 
dispersal of animals over water and through the air. Quart. Rev. 
Biol., vol. 13, pp. 274-300, ill. 

GuNDLACH, Juan 

1890-91. Contribucion a la Entomologia Cubana, T. 2, Pt. 3, pp. 289- 
396. Havana. (Facsimile Edition published by John D. Sherman, 
Jr., New York, 1922.) 

GURNEY, A. B. 

1937. Studies in certain genera of American Blattidae (Orthoptera). 
Proc. Ent. Soc. Wash., vol. 39, pp. 101-112, ill. 

1939. A revision of the Neotropical genus Xestoblatta Hebard (Orthop- 
tera; Blattidae; Pseudomopinae). Ibid., vol. 41, pp. 97-128, ill. 



GURNEY: CUBAN BLATTIDAE 59 

Hebard, Morgan 

1916a. A new genus, Cariblatta, of the group Blattellites (Orthoptera, 

Blattidae). Trans. Amer. Ent. Soc, vol. 42, pp. 147-186, ill. 
1916b. Studies in the group Ischnopterites (Orthoptera, Blattidae, 

Pseudomopinae). Ibid., vol. 42, pp. 337-383, ill. 
1917. The Blattidae of North America north of the Mexican Boundary. 

Mem. Amer. Ent. Soc, No. 2, 284 pp., ill. 
1920. The Blattidae of Panama. Ibid., No. 4, 148 pp., ill. 

1929. Previously unreported tropical American Blattidae in the British 
Museum (Orthoptera). Trans. Amer. Ent. Soc, vol. 55, pp. 345- 
388, ill. 

KiRBY, W. F. 

1903 Notes on Blattidae & c. with descriptions of new genera and 
species in the collection of the British Museum, South Kensington. 
No. 11. .\im. Mag. Nat. Hist., ser. 7, vol. 12, pp. 273-280. 

1904. A Synonymic Catalogue of Orthoptera. Vol. I. London, 501 pp. 

1910. Ditto. Vol. 3. London, 678 pp. 

QuADRi, M. A. H. 

1940. On the development of the genitalia and their ducts of orthopter- 
oid insects. Trans. Ent. Soc. Lond., vol. 90, pp. 121-175, ill. 

Rehn, J. A. G. 

1903. Studies in American Blattidae. Trans. Amer. Ent. Soc, vol. 29, 
pp. 259-290. 

1930. New or little known Neotropical Blattidae (Orthoptera). Number 
two. Ibid., vol. 56, pp. 19-71, ill. 

1932. Same title. Number three. Ibid., vol. 58, pp. 103-137, ill. 

1936. The Hispaniolan genus Polyancistrus (Orthoptera: Tettigoniidae, 

Pseudophylhnae). Ibid., vol. 62, pp. 271-316, ill. 
1937a. New or little known Neotropical Blattidae (Orthoptera). Number 

four. Ibid., vol. 63, pp. 207-258, ill. 
1937b. The Cuban genus Polyancistroides (Orthoptera: Tettigoniidae). 

Ibid., vol. 63, pp. 333-345, ill. 

Rehn, J. A. G. and Hebard, Morgan 

1927. The Orthoptera of the West Indies. Number 1. Blattidae. Bull. 
Amer. Mus. Nat. Histo., vol. 54, art. 1, pp. 1-320, ill. 

SCHTXCHERT, ChARLES 

1935. Historical Geology of the Antillean-Caribbean Region. New York, 
London, 811 pp., ill. 

Sharp, David 

1895. The Cambridge Natural History, vol. 5, 584 pp., ill. 

She.\ly, A. L. 

1927. Science (N. Y.), vol. 66, pp. 426-427. 



60 bulletin: museum of comparative zoology 

Shelford, R. 

1906. Studies of the Blattidae. Trans. Ent. Soc. Lond., 1906, pp. 231- 

278, ill. 
1908. Subfamily Phyllodromiinae, Genera Insectorum, Fasc. 73, pp. 1- 

29, ill. 
1910. Subfamily Blattinae, Ibid., Fasc. 109, pp. 1-27, ill. 

Snodgrass, R. E. 

1937. The male genitalia of orthopteroid insects. Smithsonian MisQ. 
Coll., vol. 96, No. 5, 107 pp., ill. (Fubl. 3442.) 

Taylor, Normax 

1916. The ascent of the Turquino, the highest mountain in Cuba. 
(Including letter of April 17, 1860, by F. W. Ramsden.) Torreya, 
vol. 16, pp. 211-225, 1 map. 

Tepper, J. G. O. 

1893. The lilattariae of Australia and Polyne.sia. Trans. Roy. Soc. S. 
Aust., vol. 18, pp. 25-130. 



EXPLANATION OF PLATES 



PLATE 1 



GuRNEY — Cuban Blattidae 



PLATE 1 

Fig. 1. Eurycotis rhodae, new species, female. Dorsal view. 

2. E. taunis R. & H., female. Dorsal view of latero-posterior process 
of tergum 7. Coast below Pico Turquino, Oriente Province, Cuba, 
June 26-30, 1936. 

3. E. fugacis, new species, male. Ventro-mesal view of right phallomere. 

4. E. tibialis Hebard, male. Ventral view of right hind tibia. Holotype. 

5. E. opaca (Brunner), female. Dorsal view of latero-posterior process 
of tergum 7. San Diego de los Banos, Pinar del Rio Province, Cuba, 
April 22, 1900. 

6. E. tor quinensis R. &li., male. Same view as in fig. 3. Pico Turquino 
(south side), Oriente Province, Cuba, 3,000-5,000 feet, June 1936. 

7. E. famelica, new species, female. Dorsal view. 

8. E. fugacis, new species, female. Dorsal view. 

9. E. caudellana, new species, female. Dorsal view. 



BULL. MUS. COMP. ZOOL. 



Gurney: Cuban Blattidae. Plate 1 




y.famellca afuyacis 9. caudellana 



PLATE 2 



GuRNEY — Cuban Blattidae 



PLATE 2 

Fig. 10. Eurycotis tibialis Hebard, Male. Dorsal view of supraanal plate 
Holotype. 

11. E. torqtdnensis'R. &'SL.,iemB\e.. Same view. Same data as in fig. 6. 

12. E. lixa Rehn, male. Same view (after Rehn, 1930). 

13. E.fugacis, new species, male. Same view (apex deformed). 

14. E. opaca (Brunner), female. Same view. Same specimen as in fig. 5. 

15. E. lixa Rehn, male. Dorsal view of left tegmen (after Rehn, 1930). 

16. ^. opaca (Brunner), male. Dorsal view of supraanal plate. Santiago 
de Las Vegas, Havana Province, Cuba, April 1905. 

17. E. taurus R. & H., female. Same view. Same specimen as in fig. 2. 

18. Same, male. Same view (after Rehn and Hebard, 1927). 

19. E. fugacis, new species, male. Frontal view of head. 

20. E. tibialis Hebard, male. Dorsal view of left tegmen. Holotype. 

21. E.famelica, new species, female. Same view as in fig. 19. 



BULL. MUS. COMP. ZOOL. 



Gurney: Cuban Blattidae. Plate 2 




10. tibialis 6 




14. opaca ? 



17. taurus ? 




1 1. torqainensis 



13. fugacis d 



15. Iixa 





16. opaca d" 




18. taurus cf 




19. fugacis 



20. tibialis 21. famelica 



PLATE 3 



GuRNEY — Cuban Blattidae 



PLATE 3 

Fig. 22. Carihlattoides instigator R. & H., male. \'entro-posterior view of 
subgenital plate. Gran Piedra Range, Oriente Province, Cuba, 
2,000-3,000 feet. May 30-31, 1936. 

23. Same, male. Same view. Mountains north of Imias, Oriente Prov- 
ince, Cuba, 3,000-4,000 feet, July 25-28, 1936. 

24. Aglaopteryx mira Rehn, male. Ventral view of subgenital plate. 
Loma del Gato, Cobre Range, Oriente Province, Cuba, about 3,000 
feet, July 3-7, 1936. 

25. Ischnoptera ligula collina, new species, male. Dorsal view of supra- 
anal plate. Paratype from Loma del Gato. 

26. Same, male. Dorso-posterior view of subgenital plate. Holotype. 

27. /. darlingtoni, new species, male. Dorso-posterior view, slightly from 
left side, of right style. Holotype. 

28. Same, male. Posterior view, from left side, of left paraproct. Holo- 
type. 

29. Same, male. Posterior view, from right side, of right paraproct. 
Holotype. 

30. Same, male. Dorso-posterior view, slightly from left side, of sub- 
genital plate. Holotype. 

31. Same, male. Dorsal view of apex of abdomen. Holotype. 



BULL. MUS. COMP. ZOOL. 



Gurney: Cuban Blattidae. Plate 3 




25. coHi'na 



22. iiistigator 





25. initigator 




29. darlingtoni 



28. darlingtoni 



27 darlingtoni 




30. darling ton! 




PLATE 4 



GuBNEY — Cuban Blattidae 



PLATE 4 

Fig. 32. Neoblattella guanayara, new species, male, ^'entro-posterior view of 
subgenital plate. Holotype. 

33. Pholadoblatta inusitata (Rehn), male. Ventro-posterior view of sub- 
genital plate. Soledad (near Cienfuegos), Santa Clara Province, 
Cuba, May 1936. 

34. N elipophygiis banksi, new species, male. Ventral view of subgenital 
plate. Holotype. 

35. Same, male. Dorsal view of left tegmen. Holot3T)e. 

36. Same, male. Apical view of claws and associated structures of left 
hind tarsus. Holotype. 

37. Same, female. Dorsal view of supraanal plate. Allotype. 

38. Neoblattella vatia R. & H., male. Ventral view of subgenital plate. 
Pico Turquino (south side), Oriente Province, Cuba, 1,500 feet, 
June 25, 1936. 

39. Symploce viunda, new species, male. Ventral view, considerably from 
left side, of subgenital plate. Holotype. 

40. Epilampra cubensis Bolivar, female. Dorsal view of right tegmen. 
Cuchillo de Guajimero (near Imias), Oriente Province, Cuba, about 
2,000 feet, July 25, 1936. 

41. Aiidreia exploratrix, new species, female. Dorsal view, from left 
side, of left tegmen. Paratype. 

42. Same, male. Ventral view of subgenital plate. HolotA'pe. 

43. Same, male. Dorsal view. Holotype. 

(All drawings by the author) 



BULL. MUS. COMP. ZOOL. 



Gurney: Cuban Blattidae. Plate 4 




^2. guaneiyara 



33.inLL5itatc 




41.exploratr.ix 



37. banks i 9 




40. cubensis 





39. TTiLLnda 



42.exploratrix 4Aexploratrix 



Bulletin of the Museum of Comparative Zoology 

AT HARVARD COLLEGE 

Vol. LXXXIX, No. 3 



NOTES ON VARIOUS PEREGRINE EARTHWORMS 



G. E. Gates 
Judson College, Rangoon, Bui*ma 



CAMBRIDGE, MASS., U.S.A. 

PRINTED FOR THE MUSEUM 

February, 1942 



No. Z. — Notes on Various Peregrine Earthworms 
By G. E. Gates 

TABLE OF CONTENTS p^^^ 

Introduction 64 

Family Ocnerodrilidae 65 

Genus Eukerria Michaelsen 1935 67 

Etikerria j^eguana spec, nov 67 

Eukerria saltensis (Beddard) 1895 73 

Genus Gordiodrilus Beddard 1892 75 

Gordiodrilus peguanus spec. no\' 85 

Genus Malaharia Stephenson 1924 90 

Key to species of Malaharia 92 

?MaIabaria levis (Chen) 1938 92 

Malaharia species 98 

Genus Ocnerodrilus Eisen 1878 99 

Ocnerodrilus occidentalis Eisen 1878 99 

Genus Thatonia gen. nov 101 

Thatonia gracilis spec. no\- 101 

Family Megascolecidae 107 

Genus Woodwardiella Stephenson 1925 107 

Woodwardiella javanica (Michaelsen) 1910 1C8 

Woodwardiella jmmila Stephenson 1931 112 

Genus Phcretima Kinberg 1867 119 

Pheretima hicincta (E. Perrier) 1875 119 

Pheretima humilis spec, nov 120 

Genus Ramiella Stephenson 1921 122 

Ramiella cultrifera Stephenson 1931 122 

Genus Dichogaster Beddard 1888 128 

Dichogaster affinis (Michaelsen) 1890 128 

Dichogaster holaui (Michaelsen) 1891 129 

Dichogaster modiglianii (Rosa) 1896 130 

Dichogaster saliens (Beddard) 1893 134 

Family Eudrihdae 137 

Genus Eudrilus E. Perrier 1871 137 

Eudrilus eugeniae (Kinberg) 1867 137 

Bibliography 144 



64 bulletin: museum of comparative zoology 

INTRODUCTION 

The peregrine earthworms of tropical and subtropical regions have 
not received the attention due them because of their extensive distri- 
bution. The anatomy of many of the species is inadequately character- 
ized and the taxonomic status of a number is uncertain. 

A previous article (Gates, 1937) is concerned with the synonymy, 
local distribution and definition of the Pheretimas which have been 
imported into America from the Oriental region. In another article 
now in press some of the confusion prevailing with regard to the septal 
displacements in a pantropical form has been cleared up. The portion 
of the survey of non-Lumbricid, peregrine forms included in the present 
contribution is merely in the nature of a report on the material that 
could be obtained during the last five to six years. The study of one 
Ocnerodrilid species naturally led to an examination of one of the most 
difficult problems of Megadrilid systematics, that provided by the 
genus Gordiodrilus, and several suggestions as to a solution are ad- 
vanced. Among other results of especial interest may be mentioned 
the presence in Eudrilus eugcniac (Kinberg) 1867, of supra-intestinal 
glands hitherto known only from three Megascolecid genera, and of 
"rolled tube" penial setae, hitherto recorded only from three Eudrilid 
genera, in a species of a Megascolecid genus, Ramiella. 

The tendency in the past has been to regard a species as endemic in 
the area in which it is first found, ^ in spite of recognizing a tendency to 
transportation on the part of small-sized species, limited generic repre- 
sentation in the area concerned, and absence of those large species 
which are less likely to be successfully transported. Forty years or 
more without further records of distribution seem only to confirm the 
endemicity. Hence the description of new species and especially a new 
genus (of the Ocnerodrilidae) in an article on peregrine forms may at 
first seem rather surprising. The earthworm faunae of large areas of 
the world are very incompletely worked out, particularly those of the 
African and Neotropical regions, regarded as the main areas of Ocnero- 
drilid evolution. Earthworms are known to have been transported 
widely with frequent colonizations in new areas far from their original 
homes. In view of the fragmentary knowledge of so many large areas, 
the spasmodic, casual or superficial nature of most work on megadrilid 
faunae, it would be premature to conclude that more than a portion of 
the successful transportations have been recognized hitherto. 

1 Woodwanliellas. for iiislance, havf: })een supposed to be endemic in South India and Ceylon , 
as also two species of Eukerria in Southern California, and species of Pherelima in Madagascar. 
Endemicity of three species of Howascolex and of two species of Dichogaster in South India may 
also be cited. 



gates: peregrine earthworms 65 

Admittedly transportation, as an explanation of the occurrence of 
two of the new Ocnerodrilid species in the regions in which they are 
now alone known, remains to be demonstrated. Nevertheless it is 
most probable that the new species of Eukerria is an importation into 
Burma, either directly or indirectly from South America. Endemicity 
of a species of Gordiodrilus or of Woodwardiella in a region including 
South India and Burma, or of a species of Malaharia in a region includ- 
ing Burma and Hainan Island, is highly improbable. In fact there is 
some justification for doubting endemicity of any Ocnerodrilid in any 
part of the Oriental region. Likewise there are good reasons for sus- 
pecting that the Oriental Woodwardiellas are importations from the 
Australasian region. 

In these circumstances it seems to be preferable to label a species as 
probably or even only possibly peregrine, rather than spare the label 
and confuse zoogeographical discussion, particularly for the sake of 
those who may be quite unfamiliar with the difficulties involved in 
connection with the distributions of Megadrilids. 

The author's thanks are extended to the University of Rangoon for 
grants to meet costs of Burmese collections, and to Miss Chapman for 
assistance in measuring penial setae. 

FAMILY OCNERODRILIDAE 

An anterior, pre-intestinal region of the body of earthworms always 
differs considerably from the intestinal portion. On passing from the 
intestinal to the pre-intestinal region, the function of the dorsal blood 
vessel changes from collection to distribution, while that of the ventral 
trunk changes from distribution to collection; segmental commissures 
are replaced in part at least by hearts, and other modifications of 
the vascular system are found, particularly in connection with the 
subneural system (when present) and the enteric plexi. The excretory 
system is likewise often modified in a larger or smaller portion of 
the pre-intestinal region. Calciferous tissues, often in special glands, 
are associated with the pre-intestinal region and a considerable portion 
of the gut in that region, if not all of the oesophagus is lined with 
cuticle.' The pre-intestinal region is also often characterized by an 

I Cuticular secretion seems to be characteristic of ectoderm, and accordingly portions of the 
gut derived from the stomadeal and proctodeal invaginations may be expected to be lined with 
cuticle. But in species of Pheretima the cuticular lining of the oesophagus lias been recognized 
into X, while in species of Drawida the lining has been found as far back as xxv. Although as yet 
undemonstrated in more posterior portions of the gut, there is a possibility that the cuticular 
lining extends to the hind end of the oesophagus. In these circumstances an ectodermal origin 
of the whole of the oesophagus from mouth to oesophageal valve perhaps requires considera- 
tion. An ectodermal origin for the whole of the oesophagus would be another distinguishing 
characteristic of the pre-intestinal region. If on the contrary the gut epithelium of the pre- 
intestinal region is not ectodermal, it is still distinguished from that of the intestine by its 
ability to secrete cuticle. 



66 bulletin: museum of comparative zoology 

extra-oesophageal system of blood vessels, which may have somewhat 
of a portal nature, transporting blood between the skin and the cal- 
ciferous section of the gut. 

The pre-intestinal region varies in segmental extent and may include 
only the first eleven segments as in the Ocnerodrilidae, the anterior- 
most thirteen as in certain species of Ramiella, Eudrilus, and Tono- 
scolex, fourteen as in Pheretima, Eutyphoeus and a number of other 
Megascolecid genera, fifteen as in Hoplochaetella, sixteen as in Nello- 
gaster, seventeen as in Dichogaster, Notoscolex (part) and Megascolex 
(part), eighteen as in Notoscolex (part), Priodochaeta and Priodoscolex, 
nineteen or twenty as in certain species of Diplocardia, more than 
twenty in the Moniligastridae. 

Assuming that all earthworms have had a common origin, these 
variations may have resulted from (1) maintenance or reduction of a 
primitively extensive pre-intestinal region, (2) maintenance or exten- 
sion of a primitively short region, or (3) simultaneous developments 
in different ways in different phylogenetic lines. The evidence available 
is insufficient to warrant much discussion of these possibilities at pres- 
ent. It may however be noted that there seems to be a tendency to 
extension of the pre-intestinal region in certain evolutionary lines, 
as for instance in the Megascolecinae and Diplocardia, also that Pick- 
ford regards the appearance of hearts in xiii as of recent and sporadic 
occurrence in iVcanthodriline groups in which the original condition 
was three pairs in x-xii. Now if the tendency in the Megascolecidae 
(in Stephenson's sense) is to extension rather than reduction, the usual 
phylogenetic derivation of the Ocnerodrilidae from the Acanthodrilinae, 
as the last of the stems deriving from Notiodrilus, will be incorrect. It 
may be advisable to consider if the Ocnerodrilidae are not more primi- 
tive, at least in respect to the pre-intestinal region, than the sup- 
posedly ancestral Acanthodrilinae. 

The distinguishing character of the Ocnei'odrilidae, according to 
Stephenson, is the presence of an unpaired and ventral or paired calcif- 
erous sacs in ix or ix and x. Two genera are now known without such 
sacs. Much more characteristic, on the basis of the evidence now avail- 
able, seems to be the short pre-intestinal region, with intestinal origin 
in xii and restriction of the latero-oesophageal hearts to x-xi. 

Of the 44 Ocnerodrilid species recognized by Michaelsen in 1900, 
the diameter (presumably maximum) of five species is one mm., of 
fourteen species one to two mm, , of nine species two mm., of one species 
between two and three mm., of one species three mm. Of the fourteen 
species with thickness unknown, the length is as follows: 15-38 mm. 



gates: peregrine earthworms 67 

(1 species), 15-35 mm. (2), 25 mm. (5), 25-50 mm. (1), 25-60 mm. (1), 
40 mm. (2), 40-50 mm. (1), even length unknown (1). All of these 
species of unknown diameter probably are of about the same size as 
other forms. Of forms described since 1900, the single species of 
Paulistus and the two known species of Kerriona are, so far as can be 
discovered from the literature now available, the only exceptions to 
the general rule as to size in the family, and those three species are all 
South American endemics. Probably no earthworms are smaller than 
the Ocnerodrilids. So far then as size is concerned the Ocnerodrilidae, 
of all earthworm families, appear to be most favorably adapted for 
transportation in numbers sufficient to enable colonization when con- 
ditions of the new habitat are suitable. 

In view of this small-size predisposition to transportation, it is 
necessary to be cautious in attributing endemicity to any Ocnerodrilid 
species in any area separated by sea from Africa or America, and even 
in portions of Africa itself as well as America. The occurrence of a 
supposed endemic of the "fairly archaic" Eukerria in South Africa 
was one of Michaelsen's reasons for believing that Africa rather than 
South America is the headquarters of Ocnerodrilid development. The 
South African form {gunningi) is now known to be peregrine, and 
actually is, according to Pickford, only saUensis. Other supposed en- 
demics that might better be regarded as possibly if not probably pere- 
grine are two species of Eukerria in Lower California, one species of 
Ocnerodrilus in Paraguay, species of Gordiodrilus in the Mediterranean 
region of Africa, species of Maheina in the Seychelles, species of Gurgia 
and Malabaria in India, species of Gordiodrilus in Madagascar. 

According to Michaelsen's criteria of large and small-sized species 
versus small-sized species only. South America would appear to be of 
greater importance in Ocnerodrilid development than Africa. Or is 
the situation comparable to that of Dichogaster, which crossed the 
Atlantic Ocean from Central America into Africa to give rise to the 
giant forms that enable recognition of African endemicity and major 
development? 

Genus Eukerria Michaelsen 1935 

Eukerria peguana spec. nov. 

Material examined. From Burmese collections: 

Rangoon, Hanthawaddy district, February, 3 clitellate specimens; June, 
53 juvenile or aclitellate specimens and 1 partially clitellate specimen; 
July, 89 juvenile or aclitellate specimens; August, 11 clitellate speci- 
mens. K. John. 



68 bulletin: museum of comparative zoology 

Kungyangon, Hanthawaddy district, September, 4 clitellate specimens. 
K. John. 

Thongwa, Hanthawaddy district, September, 1 clitellate specimen. 
K. John. 

Mouhnein, Amherst district, October, 2 aclitellate and 3 clitellate speci- 
mens. K. John. 

Pegu, Pegu district, August, 3 clitellate specimens. K. John. 

Wanetchaung, Insein district, September, 1 clitellate specimen. K. John. 

Myaungmya, Myaungmya district, October, 1 aclitellate, 1 partially 
clitellate and 4 clitellate specimens. Maung Ohn Maung. 

Pyinmana, Yamethin district, October, 3 clitellate specimens. K. John. 

External charncteristics. Length 20-50 mm. Diameter ^-1 mm. 
Unpigmented. Prostomium prolobous. Dorsal pores lacking. Neph- 
ropores unrecognizable. Setae begin on ii on which all four couples 
are present; ab and cd about equal, aa and be usually about equal 
though there is some variation, dd ca. = 3^C. 

The clitellum is reddish to light yellowish, annular, not protuberant, 
apparently extending from a posterior portion of xiii to 19/20 or onto 
the anteriormost portion of xx but anterior and posterior limits are 
unrecognizable externally ; intersegmental furrows lacking, setae pres- 
ent. In aa the clitellar coloration is often lacking but the epidermis is 
thickened though not as much as lateral to b, clitellar thickening lack- 
ing ventrally on xiii and probably also on xvii-xx. 

The spermathecal pores are transversely placed slits in ab, on 7/8 
and 8/9, the centres about at mid ab. The margins of the apertures 
are protuberant (occasionally wrinkled) but are not delimited periph- 
erally to form a lip. 

The female pores are minute, transversely placed slits slightly be- 
hind the site of 13/14, just lateral to b. 

On xvii-xix, on each side, and extending median to a and lateral to b 
is a longitudinally placed, slightly raised, fairly sharply demarcated, 
rather dumb-bell-shaped area that reaches to or nearly to 19/20 and 
nearly to 16/17, a central portion of the nearly circular anterior and 
posterior end parts especially protuberant. Along each of these areas 
and reaching nearly to the setal arcs of xvii and xix there is a nearly 
straight band of greyish (epidermal?) translucence. This band usually 
has a smooth rather than a grooved appearance (i.e., no trace of seminal 
groove recognizable) but on the clitellate Rangoon worms the band 
is less obvious, apparently replaced on some specimens by a slight 
groove. At the anterior and posterior ends of each band or groove and 
about on the setal arcs of xvii and xix there is a transversely placed 



gates: peregrine earthworms 69 

slit-like aperture which opens into a slight parietal invagination with a 
transversely slit-like lumen. On most of the specimens each aperture 
is occupied by a transversely placed, shortly ellipsoidal, transparent to 
translucent vesicle, the anterior vesicle attached to the posterior wall 
of the invagination, the posterior vesicle to the anterior wall. These 
vesicles may be conspicuously protuberant from the body wall on short 
but narrow necks or fully or partially retracted into the invaginations. 
After forcing the vesicle back against the wall to which it is attached 
there becomes visible on the roof of the invagination a minute pore 
which is probably the prostatic aperture. On the setal arc of xviii and 
on each longitudinal grey band there is a minute opening, probably the 
male pore. Median to this aperture and also on the grey band setae a 
and b of xviii are usually visible, slightly displaced mesially. Occasion- 
ally only one of a pair is visible. Ventral setae of xvii and xix are in\'is- 
ible or unrecognizable externally but within the parietes just median 
to the ectal ends of the prostatic ducts, after separation of the lon- 
gitudinal muscle fibres, there may be present setae, so closely paired 
that they appear to be in the same follicle (attempts at removal un- 
successful). 

The genital markings are unpaired, transversely placed areas of epi- 
dermal thickening with rounded lateral margins, reaching anteropos- 
teriorly to or almost to the intersegmental furrows, and extending 
laterally on each side into the median portion of be. The epidermal 
thickening is much more marked laterally than midventrally but there 
is no other indication of a double origin such as incisions of anterior 
and posterior margins at the midventral line. Just lateral to b on each 
marking is a minute aperture. Markings are located as follows: on xxi 
(29 specimens), on xxi — right side only (1), on xxi — left side only (1), 
on XX and xxi (4, on two of which the anterior marking is asymmetri- 
cal). One clitellate specimen has no marking. 

On juvenile and aclitellate Rangoon specimens the dumb-bell- 
shaped male porophores are lacking but on the larger specimens pro- 
static pore invaginations and associated vesicles are developed. Geni- 
tal markings and gland apertures are quite unrecognizable though the 
glands are present. 

Internal anatoiny. All septa from 5/6 posteriorly are present; 5/6- 
8/9 thickened and muscular. 

The pharyngeal bulb is small and short. The oesophagus in iv-vi 
is narrow and with thin, transparent or translucent wall. There are 
masses of iridescent glands (?) in iv-vi, those in v especially large. The 
gizzard, in ^di, appears to be more muscular than in saltensis. Calcif- 



70 bulletin: museum of comparative zoology 

erous sacs are like those of saltensis except that here the lumen is 
triangular in section (4 sacs) and the sacs do not reach to 8/9 to which 
they are connected by strands of tissue. Longitudinal ridges are visible 
on the inner wall of the sac, the wall so thick and the lumen so small 
that the walls are almost in contact centrally. The intestine begins in 
xii, the oesophageal valve distinct and in the anterior part of the seg- 
ment. No typhlosole. 

The dorsal blood vessel is single. The supra-oesophageal trunk ex- 
tends from a posterior portion of viii into xi where it bifm'cates just in 
front of 11/12, the branches passing laterally onto the anterior face of 
11/12 behind the hearts and then ventrally onto the parietes where 
they can be traced anteriorly into ix. In one specimen, on the left side, 
a large vessel rises from the parietes onto the anterior face of 10/11 and 
then passes behind the heart of x to open into the supra-oesophageal 
trunk. In ix a large vessel from the dorsal face of each calciferous sac 
joins the supra-oesophageal just anterior to 9/10. Extra-oesophageal 
trunks are recognizable just ventrolateral to the gut in vi, running 
postei'iorly through vii and viii, in ix passing onto the ventral faces of 
the oesophageal sacs where they become invisible. The hearts are 
two pairs, in x-xi, dorsally an anterior branch from each heart opening 
into a bifurcation of the supra-oesophageal in xi, or into the vessel from 
the parietes in x, or directly into the supra-oesophageal, a posterior 
branch opening into the dorsal blood vessel. The commissures of ix 
may be as large as the hearts of x and xi but only open into the dorsal 
trunk from which smaller commissures are given off in viii-v. Ventral 
stumps of the commissures of vii-ix and of the hearts of x-xi join the 
ventral trunk but the vessels have not been traced completely around 
the oesophagus. The ventral trunk is unrecognizable anterior to vii. 
No subneural trunk. 

The nephridia from xx posteriorly are large, in contact with both 
septa of a segment, extending from mid he nearly to the mid-dorsal 
line. The preseptal funnels are delicate, translucent, small, just median 
to a and close to the ventral parietes, the neck long, very slender and 
transparent. In xix-xii the nephridia are much smaller and in the 
anterior portions of the segments just behind the septa (funnels not 
found). From xi anteriorly nephridia have not been found and if pres- 
ent must be still smaller. 

The coelomic cavity of x is filled with a loose, non-iridescent coagu- 
lum. The funnels (one pair only) are large, frilled, and characterized 
by a brilliant pinkish iridescence. The posterior vesicles are large, 
reaching up into contact with each other above the dorsal blood 



gates: peregrine earthworms 71 

vessel, lateromesially flattened, pushing 11/12 back into contact with 
12/13 (1) or penetrating 11/12 and dislocating 12/13 posteriorly. The 
anterior vesicles are large, reaching up into contact with the dorsal 
trunk and filling all of the available space in ix. Both pairs of vesicles 
are lobed, those of ix softer and more fragile. The prostates are at 
least six mm long, reaching back into xxviii-xxx but looped and in part 
twisted together so that they are long enough to reach several segments 
further. The ducts are 3^-1 mm long, much slenderer than the glands, 
opaque but with no muscular sheen, passing into the parietes in xvii 
and xix. The vas deferens passes into the parietes in xviii just lateral 
to a pair of setae. 

The sphermathecal ampulla is shortly ellipsoidal, filled with a non- 
iridescent whitish or yellowish material. The coelomic portion of the 
duct is about as long as the ampulla, slightly narrower, rather flat- 
tened, with slight constrictions (2-4) that produce a rather irregularly 
moniliform appearance, bound down to the parietes. The wall is 
slightly thicker than that of the ampulla, the lumen transversely slit- 
like to flatly elliptical in cross section. In the parietes the duct is 
somewhat narrowed and the wall is tougher but the lumen is still 
transverse. In an ectal portion of the duct of each of several sperma- 
thecae there is an elongate, rather pyriform mass of iridescent material, 
presumably spermatozoa. 

Each ovary contains only a few relatively large ova. 

The genital marking glands are stalked and coelomic, tubular. The 
duct is of about the same length as that of a prostate but is slenderer, 
translucent, and sinuous. The gland is one to three times as long as 
the duct, 3^-13^ mm long, much slenderer than the prostates and 
of course shorter. A delicate connective tissue investment is recog- 
nizable at the margin of the gland. In sections the epidermis of the 
genital markings appears to be about half again as thick as in the sur- 
rounding region. 

The epidermis is also considerably thickened at the anterior and 
posterior ends of the dumb-bell-shaped male porophores. The pro- 
static duct passes into the centre of a fairly sharply demarcated area of 
circular outline in the parietes but this area is not protuberant into the 
coelomic cavity. In xvii-xix there are numerous diagonal muscle fibres. 

Remarks. Although anterior and posterior boundaries of the clitel- 
lum are unrecognizable externally in the sectioned body wall at the 
mid-dorsal incision, the clitellar thickening of the epidermis is clearly 
visible even on xiii and xx, 

Spermathecal apertures of the Kungyangon and a few other speci- 



72 bulletin: museum of comparative zoology 

mens are open and occupied by plugs of transparent material that are 
continued internally for some distances into the coelomic portions of 
the spermathecal ducts. 

A middle portion of the dumb-bell-shaped male porophore may be 
scarcely recognizable or quite invisible. The longitudinal grey lines 
may be bent laterally very slightly in the region of the setal arc of 
xviii or in such a way as to subtend a very wide, obtuse angle. 

The transparent vesicle of the prostatic pore invagination appears 
to be similar to the "clear gland" associated with the ectal end of the 
prostatic duct of Aphanascus oryzivorus Stephenson 1924. Stephenson 
(1924, p. 362) mentions a possibility that the "clear glands" disappear 
at full maturity. No evidence for such disappearance is provided by 
the Burmese specimens of E. peguana. 

The left calciferous sac of one Moulmein specimen is lacking, the 
right gland about twice the usual size. 

E. peguana, with spermathecal pores in ab, has affinities with the 
following species: E. halophila (Beddard), 1892, E. asuncionis (Rosa) 
1895, E. Jdlkenthali (Michaelsen) 1908 and E. selangorensis (Stephen- 
son) 1930. From each of these species peguanus is distinguished by the 
presence of "clear glands" in prostatic pore invaginations, presence of 
genital markings and the prostate-like glands associated with the 
genital markings. 

Diagnosis. Prostatic pores on roofs of transversely slit-like parietal 
invaginations with transversely placed apertures in ah on protuberant 
anterior and posterior ends of paired, longitudinally placed, dumb-bell- 
shaped porophores; a transversely ellipsoidal, protrusible clear gland 
on the posterior wall of each anterior invagination and the anterior wall 
of each posterior invagination. Male pores on setal arc of xviii just 
lateral to h which is slightly displaced mesially. Genital marking a 
transversely placed area of epidermal thickening on xxi, reaching 
anteroposteriorly nearly to intersegmental furrows and laterally into 
he, a pore just lateral to h on each side. Quadrithecal, spermathecal 
pores in ah, with centres about at mid ah. Setae: ah = cd, aa — he, 
dd = 3^C; ventral setae of xvii and xix present (?) but unrecognizable 
externally. Nephropores? Clitellum annular but thinner and colorless 
in aa; on xiii-xx. Prostomium prolobous. Unpigmented. Length 20- 
50 mm. Diameter 3^-1 mm. 

Gizzard in vii. Seminal vesicles in ix and xi. Spermathecal duct 
about as long as ampulla and nearly as wide but with slightly thicker 
wall and transversely slit-like lumen, moniliform and bound to the 
parietes. Genital marking glands tubular, stalked and coelomic. 



gates: peregrine earthworms 73 

Distribution. Known at present only from Burma (Amherst, Han- 
thawaddy, Myaungmya, Insein and Yamethin districts) but possibly 
as widely distributed in the tropics as E. saltensis and other peregrine 
Ocnerodrilids. 

EuKERRiA saltensis (Beddai'd) 

1895. Kerria saltensis Beddard, Proc. Zool. Soc. London, 1895, p. 225. (Type 

locality Salto, Valparaiso, Chile. Types in the British Museum.) 
Material examined. From Burmese collections: 

P\nnmana, Yamethin district, October, 1 clitellate specimen. K. John. 
Monywa, Lower Chindwin district, September, 1 clitellate specimen. Saw 
San Thwe. 

External characteristics. Length 27-30 mm. Diameter ca. 1 mm. 
Unpigmented. Prostomium epilobous but with no furrow at posterior 
margin of tongue. Dorsal pores lacking. Nephropores unrecognizable. 
Setae begin on ii on which all four couples are present; ab and cd about 
equal, aa slightly larger than or about equal to be. 

The clitellum is light brownish, saddle-shaped (?), not protuberant, 
reaching ventrally to b, extending antero-posteriorly onto xiii and at 
least to 19/20 but without recognizable anterior and posterior demarca- 
tion ; intersegmental furrows lacking, setae present. The ventral setae 
of xiv-xvi appear to be enlarged. 

The spermathecal pores are not minute, being definitely larger than 
other reproductive apertures, shortly elliptical, on 7/8 and 8/9, 
slightly median to c. The margin of each aperture is tumescent as a 
clearly demarcated, annular lip. 

The female pores are on b, slightly behind the site of 13/14, the 
margins slightly tumescent so that the pores appear to be transversely 
slit-like. 

In the male region, on each side, there is a longitudinally placed, 
slightly raised, rather dumb-bell-shaped area just lateral to b, extend- 
ing anteroposteriorly to just in front of or just behind the setal arcs of 
xvii and xix. The anterior and posterior portions of each porophore 
are especially protuberant, a midventral region between the poi'ophores 
slightly depressed. Along each porophore there is a greyish translucent 
band on which no seminal groove is visible (possibly grooved on Pyin- 
mana specimen?). Immediately behind the setal arc of xvii (Monywa 
specimen) there is in the epidermis a transversely placed, slightly 
crescentic band of greyish translucence, the concave side posteriorly. 
The minute prostatic pores are at the ends of the longitudinal grey 
bands of the porophores, the male pores also in the bands, on xviii, on 



74 bulletin: museum of comparative zoology 

the setal arc (all pores just lateral to 6). Ventral setae of xvii-xix 
are all present but probably are smaller than on neighboring seg- 
ments. 

Internal anatomy. (Monywa specimen). All septa from 5/6 pos- 
teriorly are present; 6/7-8/9 thickened and opaque but with no mus- 
cular sheen. 

The pharyngeal bulb is short. At each side of the gut in iv, v and vi 
there are masses of glandular (?) material with a brilliant iridescence, 
those in iv especially large. From the pharyngeal bulb to 6/7 the gut 
is slender, unconstricted septally and with thin wall. The portion of 
the gut in vii has a smooth surface, rounded shape, and is fairly strong 
though not greatly muscularized. In ix the gut is quite slender and 
fairly high in the coelomic cavity. Attached to each side posteriorly 
and dorsolaterally is a rather pear-shaped calciferous sac which passes 
below the level of the gut and then anteriorly to the posterior face 
of 8/9 to which it is adherent (segment ix unusually long). The aper- 
ture into the gut is small and rather star-shaped but the lumen of the 
sac is horizontally slit-like in transverse section. The intestine begins 
in xii. No typhlosole. 

Hearts of x-xi all open into the supra-oesophageal and dorsal trunks. 
The commissures of ix are slenderer than the hearts and apparently 
open only into the dorsal blood vessel. No subneural trunk. 

Nephridia are rather small, transversely placed against the parietes, 
extending from b to just lateral to d, the neck long and slender, the 
preseptal funnel close to the ventral parietes near b. On each post- 
clitellar nephridium there is a little granular, transparent material. 
From xvi-x the nephridia are still smaller but were not seen from ix 
anteriorly. 

In the ventral portion of x there is on each side a compact iridescent 
mass of spermatozoa adherent to the male funnel from which it can be 
removed only with some care, the funnels with a brilliant iridescence. 
Testes were not seen. The vas deferens is large in xi-xiii and slightly 
iridescent throughout so that it can be traced easily in spite of the 
slenderness of the posterior portion, passing into the parietes in xviii, 
just lateral to the ventral setae. The seminal vesicles of ix are fairly 
large, filling all available space in the segment and reaching up into 
contact with the dorsal trunk. The posterior vesicles are lateromesi- 
ally flattened, reaching into contact with the dorsal blood vessel, and 
extending through 11/12 to 12/13 which is pushed into contact with 
13/14. Prostates are twisted together but are long enough to extend 
through several segments. The ducts are much thinner than the glands 



gates: peregrine earthworms /o 

and with very slight muscular (?) sheen, passing into the parietes just 
lateral to the ventral setae of xvii and xix. 

Each ovary contains a few, relatively large ova. 

The spermathecal ampulla is shortly ovoidal, the narrowed portion 
ectally. The ampulla is filled with a compact mass of spermatozoa, a 
slight iridescence visible through the translucent wall. The duct is 
slightly shorter than the ampulla and slender. There are no diverticula 
or seminal chambers. 

Remarks. The saddle-shaped appearance of the clitellum may be due, 
as in peguana, merely to a lack of clitellar coloration midventrally. 
Some confusion has resulted from attempts to use clitellar characteris- 
tics for specific definition in small species of Dichogaster as well as 
certain Ocnerodrilids. Even in larger forms where clitellar characteris- 
tics are determinable with more certainty there may be some variation 
with respect to extent and apparent shape of the clitellum. 

Diagnosis. Prostatic pores minute and superficial, on setal arcs of 
xvii and xix and at anterior and posterior ends of grey, translucent 
bands on paired, longitudinally placed, dumb-bell-shaped porophores 
just lateral to b; male pores on setal arc of xviii and on grey bands. 
Quadrithecal, spermathecal pores slightly median to c. Setae : ab = cd, 
aa ca. = be, dd = 3^C; ventral setae of xvii-xix present. Nephropores? 
Clitellum annular (but thinner and colorless ventrally?) ; on xiii-xx. 
Prostomium epilobous. Unpigmented. Length 25-30 mm. Diameter 
1 mm. 

Gizzard in vii. Seminal vesicles in ix and xi. Spermathecal duct 
slightly shorter and slenderer than the shortly ovoidal ampulla. 

Distribution. Probably widely distributed throughout tropical and 
possibly subtropical regions but recorded hitherto only from Chile, 
South Africa, Australia, New Caledonia and Burma (central region, 
Yamethin and Lower Chindwin districts). 

Genus Gordiodrilus Beddard 1892 

This genus provides one of the most difficult problems in Megadrilid 
taxonomy. In his 1930 monograph Stephenson characterized Gordio- 
drilus as follows: "Condition of male apparatus very various — in- 
completely or irregularly microscolecine, or showing a tendency to 
balantine reduction or megascolecine. Two or one or no gizzards. A 
single oesophageal sac, or a pair, in ix. Holandric, metandric, or pro- 
andric. Spermathecse without diverticula on the duct, but sometimes 
with evaginations at the ectal end of the ampulla." (The Oligochaeta, 



76 bulletin: museum of comparative zoology 

1930, p. 863). Obviously this is merely a statement of variation in a 
group of species without morphological or geographical unity and as a 
"diagnosis is so indefinite as to be meaningless", as is admitted by 
Stephenson who adds, "I can do nothing with this heterogeneous group 
of species, extending over nearly the whole African continent" (I.e. 
p. 863). 

Casual comparison of the definition above with those of other Ocner- 
odrilid genera in the same work, suggests that the Gordiodrilid chaos 
resulted, in part at least, from neglect to continue to use those charac- 
teristics which were employed, with apparent success, in definition of 
other genera in the family (in particular characteristics of the digestive 
system relating to gizzards and calciferous sacs). 

A genotype, so far as can be discovered from the literature available, 
has never been designated. One candidate for the honor is G. tenuis 
Beddard 1892 although this form is probably a ghost species, unem- 
bodied in a material type. (The single type was sectioned but whether 
sections of this as well as of other holotypes have been preserved and 
are now of use is unknown.) Some characteristics of specific, possibly 
even of generic importance are unknown but the species does lack a 
gizzard, does have a single, ventral calciferous sac in ix and four pros- 
tates, a combination of characteristics at present capable of distinguish- 
ing a genus from all other Ocnerodrilids as well as from certain groups 
of species now included in Gordiodrilus. G. tenuis is supposedl^^ metan- 
dric but mention is made of seminal vesicles in x-xiii. Vesicles are 
hardly to be expected in xi of a metandric species, and it is quite 
possilile that testis sacs of the Ocncrodrilus type, in x and xi (or even 
masses of coagulum) were mistaken for seminal vesicles. An anterior 
pair of male funnels may have been lost (as in some sections just ex- 
amined) or unrecognized. Furthermore it should be noted that metan- 
dry may appear in a group of related species without necessity or even 
possibility of generic or subgeneric division (as in Eutyphoeus, and vide 
also subsequent remarks on Malabaria). Location of prostates in xx 
and xxi rather than in xvii and xix as in the supposed ancestral form 
is evidence for a derived or specialized rather than a primitive status 
and if tenuis is actually metandric, less specialized species in a tenuis 
group might well be holandric. In spite of the posterior location of the 
male terminalia an approximation to an acanthodriline condition is 
maintained, as the male pores apparently are midway between the 
prostatic pores. A tenuis group, as will be seen below, can be defined 
in terms of maintenance of this approximation to an acanthodriline 
condition but not with respect to a normally acanthodriline segmental 



gates: peregrine earthworms 77 

location, a development somewhat similar to that in the North Ameri- 
can Megaseolecid genus Diplocardia, where the male pores may be on 
xviii, xix, xx or xxi, with corresponding shifts in the four prostates. 

A generic definition for a tenuis group of species may then be worded 
somewhat as follows. Quadriprostatic, prostatic pores at termini of 
seminal grooves, the male pores also in the grooves and intermediate^ 
between the prostatic pores (or between the first and last pairs when 
three pairs are present). Setae closely paired. Clitellum annular.^ No 
gizzard, calciferous sac ventromedian in ix. Holandric; seminal vesicles 
in (ix?) xii.3 

No mention is made in this definition of the intestinal origin in xii, 
absence of a typhlosole, restriction of latero-oesophageal hearts to x-xi, 
or the paired presetal female pores on or close to h lines of xiv, as there 
is evidence to indicate that these characteristics are found in a consid- 
erable portion if not a very large majority of the Ocnerodrilidae. Pres- 
ence of small chambers in or on the ducts of the spermathecae may 
characterize a definite group of species or even a greater proportion of 
the genus than is now recognized. "Evaginations" of spermathecal 
ampullae may be only accidental constrictions. 

In a generic group as defined above there would be included : tenuis 
Beddard 1892, elegans Beddard 1892, zanziharicus Beddard 1894, 
madagascariensis Michaelsen 1907, trarancorensis Michaelsen 1910, 
habessinus Michaelsen 1913, paski Stephenson 1928, unicus Stephenson 
1931, wemanus Michaelsen 1937, pcguanus sp. nov., and possibly also 
the following, — dithcca Beddard 1892, dominicensis Beddard 1892, 
papiUatiis Beddard 1901, mobiicanus Cognetti 1907 and chuni Michael- 
sen 1913. 

In accordance with orthodox theory as to Oligochaete evolution, the 
most primitive species of the restricted Gordiodrilus is unicus with 
normally acanthodriline male terminalia. This form is known only 
from Bhamo in northern Burma but a short distance from the Chinese 
frontier. It is improbable that this is the original home of the species. 
A longitudinal area in east Africa apparently characterized by group 2 
crosses a transverse area possibly characterized by group 3 in the region 
of Mt. Ruwenzori where the original home of unicus may perhaps be 
anticipated. Only one specimen of unicus is known and this has a 

' Supposed union of male and prostatic pores on xviii in triivancorensis needs confirmation. 
2 The clitellnm is said to be saddle-shaped in both tenuis and elegans but a figure of each 
species apparently shows clitellar development ventrally. 

2 Vesicles, recorded on several occasions from x and xi. may again be only testis sacs of the 
Ocnerodrilus type, at height of sexual development. Seminal vesicles would then be unnecessary 
and perhaps first lost in ix as apparently in some other genera of the family. 



78 bulletin: museum of comparative zoology 

single prostate in xviii, presence of which may be an abnormahty. 
However prostates are present in xviii in certain Ocnerodrilids with a 
frequency to indicate a possibility at least that the ancestral condition 
was sexprostatic rather than quadriprostatic as hitherto assumed. 

A group of species (2, unicus and any related species being 1) includ- 
ing zanzibaricus (Zanzibar), habessimis (Abyssinia), jjaski (Tangan- 
yika), icemanus (Kenya), peguaniis (Burma) and possibly also doviini- 
censis (Dominica but via Kew) and chuni (Rhodesia), has prostates in 
xvii and xviii and male pores on or near to 17/18. This condition may 
be derived from the sexprostatic merely by elimination of the prostates 
of xix and dislocation anteriorly of the male pores to points midway 
between the remaining prostatic apertures. The male pores of zanzi- 
baricus according to Beddard are "precisely" as in elegans but pores on 
elegans are at 18/19 while seminal grooves of zanzibaricus are on xvii 
and xviii. Assuming that Beddard was correct with regard to the loca- 
tion of the seminal grooves, it is unlikely that the pores are on 18/19, 
although Beddard confirmed the location of the pores of elegans in 
1901 (p. 364). Information available from distribution indicates that 
the statement as to segmental location of the seminal grooves of zanzi- 
baricus is to be accepted rather than that as to location of the pores. 

The type of chuni is abnormal and the absence of a second pair of 
spermathecae may also be an abnormality. In the description of 
dominicensis prostatic pores are at first said to be on xviii and xix 
with ventral setae of xviii and xix lacking) while on p. 95 the pores are 
said to be on xvii and xviii, a location which was confirmed by Bed- 
dard in 1901 (p. 364). The athecal condition may be an abnormality 
as in Burmese species of Malabaria. At present the original home of 
this group would appear to be somewhere in eastern Africa between 
Rhodesia and Abyssinia. 

A third group of species, including travancorrnsis (South India), 
madagascarirnsis (Madasgacar), elegans (supposedly from Lagos but 
actually found at Kew) and possibly also mobucanus (Mt. Ruwenzori), 
has prostates in xviii and xix and male pores on or near to 18/19. This 
condition may be derived from the sexprostatic by suppression of the 
prostates of xvii and the dislocation posteriorly of the male pores. 
G. mobucanus is said to be bithecal but the species is so inadequately 
characterized as to be unrecognizable. Available information is suffi- 
ficient to indicate nothing more than a possibility that group 3 is 
endemic in a more western portion of Africa than group 2, possibly 
from Mt. Ruwenzori to Lagos (Indian and possibly Madagascar record 
of peregrine forms). 



gates: peregrine earthworms 79 

G. papillatus (from Lagos) with no gizzards and a single calciferous 
sac is mentioned here because of a remote possibiHty that it provides a 
transition between groups 3 and 4. A part of the difficulty in placing 
the species is due to another of Beddard's contradictions. On p. 360 
the prostatic pores are said to be at the termini of the seminal grooves 
on xviii and xix which would indicate inclusion in section 3, but on p. 
362 the prostatic pores are said to be on xix and xx with the male pores 
in the region of 19/20. The species is however characterized by the 
presence in the male terminalia of large muscular bulbs similar to those 
of Nannodrilus. Inclusion in Namiodrilus would require an assumption 
that gizzards in vii and viii and prostates opening into the copulatory 
chambers had been overlooked. The seminal grooves present here are 
lacking in Nannodrilus. A possibility of independent development of 
copulatory chambers in non-Nannodrilid forms may have to be con- 
sidered. Copulatory chambers, though of a much smaller sort, are 
now known from one species of Eukerria and coelomic copulatory cham- 
bers apparently have evolved in different groups of the Megascolecid 
genus Phereiima. 

The fourth section contains only one species, temds (supposedly 
from Assaba, West Africa, but actually from Kew). In this species the 
prostatic pores are on xx and xxi with the male pores on or near to 
20/21. The finding of this species towards the western margin of a 
group 3 area, together with a possible Lagos origin of papillatus, is of 
interest though perhaps of no particular significance. 

In Diplocardia with developments somewhat similar to those charac- 
teristic of tenuis the word usually used is "shift", as if the prostates 
originally present in xvii and xix had been translocated posteriorly, 
perhaps as a result of a long series of very slight backward shifts or of 
a lesser number of more marked mutations. The occasional presence of 
prostate-like glands in the Ocnerodrilidae behind the usual series of 
two or three pairs, suggests an alternative, ic, the appearance of one 
or more pairs of supernumerary structures throwing the primitive re- 
lationship out of balance with a consequent posterior dislocation of the 
male pores to restore the balance. Loss of one or more prostates at the 
anterior end of the series would again upset the balance and be followed 
by still further posterior dislocation of the male pores. In this connec- 
tion it may be noted that a number of interesting problems in earth- 
worms in connection with conditions apparently suggestive of induc- 
tion, attraction or organ balance effects await investigation. 

G. ditheca (Lagos via Kew) cannot be placed in an\' group. Male 
pores are supposedly on xviii and there is a single pair of prostates but 



80 bulletin: museum of comparative zoology 

the location of the prostatic pores is unknown. Supposedly bithecal 
the location of spermathecal apertures is also unknown. The bithecal 
condition may be an abnormality as well as the absence of a second 
pair of prostates. 

N annodrilus Beddard 1894 was united with Gordiodrilus in 1913 by 
Michaelsen, who had come to believe that the male genital terminalia 
in Kerria (now Eukerria), Gordiodrilus and Nannodrihis formed a 
series that should not be interrupted, while gizzards and copulatory 
chambers were regarded as not providing characteristics of generic 
value in this particular connection. Kerria however was not united 
with Gordiodrilus and the reasons advanced do not appear to be of 
much importance. Like Gordiodrilus, Nannodrilus has only a single 
ventral calciferous sac in ix but this in itself is no more justification for 
union of the two genera than the presence of a single pair of sacs in 
ix is for a union of Eukerria, Kerriona, Haplodrilus and Nematogenia. 
A Nannodrilid group of species can be defined generically as follows: 
Male pores on xviii. Bithecal, spermathecal pores on 7/8. Clitellum 
annular. No seminal grooves. Gizzards in vii and viii. Calciferous 
sac ventromedian in ix. Holandric, seminal vesicles in ix and xii. Vasa 
deferentia of a side after union open into a thickwalled and muscular 
"copulatory chamber" into which one pair of prostates also opens. 
Spermathecae adiverticulate. This definition certainly appears to be as 
good as any now available in the Ocnerodrilid family. 

To be included in Nannodrilus are: africanus Beddard 1894 (Lagos), 
staudei Michaelsen 1897 (Egypt), phreori/ctes Michaelsen 1903 (Cam- 
eroons), schubotzi Michaelsen 1915 (French Congo) and possibly 
togoensis Michaelsen 1913 (Togo), as well as another form hitherto 
unrecognized as specifically distinct. The distribution may indicate 
an endemic area in Dahomey to French Congo, the Egyptian record a 
result of transportation. 

Here again it seems possible to start with a sexprostatic condition 
and differing from the primitive Gordiodrilid condition only in the pres- 
ence of a pair of copulatory chambers into which both pairs of male 
deferent ducts and the prostates of xviii open. From such a primitive 
stage, schubotzi (in part only) differs merely in the location of the an- 
terior prostatic pores on or near to 17/18 rather than midsegmentally 
on xvii and the location of the posterior prostatic pores also in front 
of the setal arc. Some of the types of schubotzi have a second pair of 
prostates opening into the copulatory chambers but with no external 
prostatic apertures anteriorly. Michaelsen apparently assumed that 
the anterior prostates had become disengaged from their parietal 



gates: peregrine earthworms 81 

attachment in the region of 17/18 and reattached to the copulatory 
chambers. That such a development can occur as an intraspecific 
variation seems rather doubtful and the specimens with two prostates 
on each chamber should perhaps be regarded as specifically distinct. 
In staudei the anterior prostates are on a presetal portion of xviii and 
perhaps furnish in this respect a condition intermediate between 
schubotzi and the species of the excluded specimens. In N. africanus^ the 
posterior prostates have been lost but the anterior prostates are re- 
tained in the original position on xvii. In phreorydes both anterior and 
posterior prostates have been lost. It will be interesting to learn if 
some undescribed species has retained the posterior prostates after loss 
of the anterior glands. 

G. togoensis can be included in Nannodrilus if gizzards in vii and viii 
were overlooked and would then differ from the excluded types of 
schubotzi in the absence of prostates with external apertures. If how- 
ever gizzards were actually lacking in togoensis recognition of another 
genus must be considered. This would be distinguished from Gordio- 
drilus by the absence of seminal grooves and of prostates with external 
apertures, and from Nannodrilus by absence of gizzai'ds and of pros- 
tates with external apertures. 

Relationships of the quadrithecal (cf. footnote on africanus) johneni 
Michaelsen 1936 (Belgian Congo) are even more uncertain, primarily 
because of doubt as to the calciferous glands which are said to be not 
wholly unpaired, but with two rounded "Kuppen", accurate charac- 
terization impossible because of gaps in serial sections. If an unpaired 
Nannodrilid sac is present and gizzards were overlooked, inclusion in 
Nannodrilus is possible and the predicted stage without anterior pros- 
tates is available. If gizzards are actually lacking as well as in togo- 
ensis, then both of those species might belong to a genus paralleling 
Nannodrilus in the evolution of the male terminalia. 

G. robustus Beddard 1892, with a gizzard in viii and possibly a rudi- 
mentary gizzard in vii, in accordance with procedure followed hitherto 
must be excluded from Gordiodrilus. The presence of seminal grooves 
and absence of copulatory chambers seem to make consideration of 
relationships with Nannodrilus unnecessary. Again a new generic 
group seems to be indicated. The calciferous gland of rohustv^ is at 
first said to be single and ventral but whether this was determined 

• Again we have to do with contradictory statements in Beddard's descriptions. In the 
definition of africanus the species is said to be quadrithecal but on p. 390 only one pair of 
spermathecae is mentioned and that in vii, the location of the pores not given! Further, in the 
definition of Nannodrilus the anterior pair of prostates is said to open in common with the 
deferent ducts, but fig. 3 contradicts this. 



82 bulletin: museum of comparative zoology 

from the dissection or from sections is not clear. In a footnote on p. 83 
Beddard says that a specimen dissected after completion of the account 
above has paired calciferous sacs! The material studied by Baldasser- 
oni and Michaelsen apparently indicates that the unpaired condition 
should be accepted as correct, in which case the specimen with paired 
calciferous sacs must be referable to a distinct genus (vide Diaphoro- 
drilus). W\ of the forms to be considered with rohustus are character- 
ized by presence of genital markings. Nothing is known as to parietal 
modification or glands associated with those markings but such glands 
might also provide characteristics of generic value. A high, lamellar 
type of typhlosole has been recorded from one of the forms. In view 
of the apparent rarity of that structure in the Ocnerodrilidse this might 
well be a characteristic of generic importance. A rohustus group of 
species could then be defined generically as follows: prostatic pores 
at termini of seminal grooves which also include male pores, midway 
between the prostatic pores. Genital markings present behind the 
male genital region. Setfe paired. Clitellum saddle-shaped. Gizzard in 
(vii? or?) viii; calciferous sac ventromedian in ix; high lamellar typhlo- 
sole present from xvii (to ?). Holandric, seminal vesicles in xi and xii. 
Spermathecae adiverticulate and with unusually long ducts. ^Genital 
marking glands? Included in addition to rohustus (supposedly from 
Lagos but actually from Kew) are: pampaninU Baldasseroni 1913 
(Tripoli tania) and congicus Michaelsen 1936 (Belgian Congo). The 
area of endemicity appears to be Central Africa while intervention of 
the Sahara dessert suggests transport of one species to Tripolitania. 
A sexprostatic form is unknown but presence of prostates in xviii of 
all species once more suggests a possibility of a sexprostatic ancestral 
or primitive type. Male pores have not been identified definitely in any 
of the forms but are thought to be in the region of 18/19 in pampaninU 
with posterior prostates. By analogy with Gordiodrilus, forms with 
anterior prostates should have male apertures on or near 17/18 (also 
indicated by Beddard's fig. 5) and evolution of similar intra-generic 
groups may be expected. 

Along with Nannodrilus, Michaelsen united Diaphorodrilus Cog- 
netti 1910 with Gordiodrilus as presence of three pairs of prostates and 
a location of the male pores of 16/17 are inadequate for generic dis- 
tinction, and further that the third pair of spermathecae is of no sig- 
nificance being merely in correlation with the three pairs of prostates 
with external apertures. The validity of the criticism as to the taxo- 
nomic value of the three pairs of prostates and spermathecae may be 
accepted, at least for the time being, but it is doubtful if equality in 



gates: peregrine earthworms 83 

number of spermathecal and prostatic pores is of any particular signifi- 
cance. Correlations between number of spermathecal pores and male 
pores (Stephenson) or prostatic pores (Michaelsen) have been as- 
sumed, but a hasty survey of various families of earthworms indicates 
that any equality in number of spermathecal pores with either male or 
prostatic apertures is only casual. 

DiaphorodrUus does however have a pair of lateral calciferous sacs 
in ix which seems to provide much more of an argument for union with 
Eukerria than any other characteristic pro\'ides for union with Gordio- 
drilus. The distribution of Eukerria (restricted to southern South 
x\merica, except for zonalis and mcdonaldi which are known only from 
Lower California and should be under suspicion of transportation, 
though usually regarded as endemics) presumably was the reason for 
Michaelsen's failure to follow his own suggestion (in spite of the 
presence of endemics of the subgenus Illyogenia in both Africa and 
America). Leaving aside for the present discussion of possible rela- 
tionships with other genera ha\ing paired sacs in ix only, which would 
require consideration of all of the remaining African Ocnerodrilids, 
recognition of DiaphorodrUus as generically distinct from other Gordio- 
drilids is necessary. By analogy with Gordiodrilus and Nannodrilus 
the sexprostatic genotype should be primitive. Unfortunately the 
segmental locations of the terminalia are uncertain because of another 
of the contradictions that characterize so much of the work on Gordio- 
drilids. According to one statement and one figure the prostates are 
in xvi, xvii and xvm while according to another statement and figure 
the last pair of prostates is in xix. The male pores were unrecognizable 
but Cognetti thought they were on 16/17, certainly a most unusual 
location either from the standpoint of location in the ]Megascolecidae 
(in which the Ocnerodrilidae are usually included) or from that of 
interorgan balance. Quite possibly the confusion as to location of the 
prostates was due to some sort of segmental abnormality in the region 
of the male terminalia with an apparent dislocation of male apertures. 
In that case the normal condition might well be that of the usual primi- 
tive form with male pores and apertures of a middle pair of prostates 
on xviii. If otherwise, doriae (Fernando Po) might have been derived 
from the primitive type by a mutational addition of an extra pair of 
prostates at the anterior end of the series rather than at the posterior 
end as in Gordiodrilus, followed b\' suppression of the prostates of xix 
and anterior dislocation of the male pores. All of the species of 
Gordiodrilus not hitherto mentioned agree with Biaphorodrilus in the 
possession of paired calciferous sacs in ix, adiverticulate spermathecae, 



84 bulletin: museum of comparative zoology 

a constant segmental location of spermathecal pores (rather than inter- 
segmental)/ holandry (seminal vesicles in xi-xii?). Otherwise this 
group of species at present does not appear to have much in common 
although here as elsewhere information is lacking with regard to a 
number of characteristics that may be of taxonomic value. In luyker- 
leni Michaelsen 1913 (Rhodesia) the male pores are on xviii, either 
united superficially with or immediately lateral to the prostatic pores, 
the anterior prostates suppressed. In overlaeti Michaelsen 1936 
(Belgian Congo) the male pores are united with the prostatic pores of 
xix, prostates of xvii and xviii eliminated but prostate-like glands in 
XX are associated with genital markings somewhat as in E. peguana. 
In thomseni Michaelsen 1933 (Southwest Africa) the male pores 
apparently are united with the prostatic pores of xviii while prostates 
associated with genital markings or porophores may or may not be 
present on xix. If the group is a natural one developments would 
appear to be along lines of loss of acanthodriline relationships and of 
seminal grooves and the evolution of megascolecine and balantine 
terminalia or approximations thereto. 

The suggestions advanced above may be expected to stimulate a 
natural but superficial criticism of unnecessary generic multiplication. 
It should however be noted that Africa is the region in which the Oc- 
nerodrilid family "took its rise" (Stephenson 1930, p. 853) and pre- 
sumably greatest development. Yet only four genera are now recog- 
nized from the whole of the African continent. Furthermore the 
generic distinctions that have been suggested above are on the whole 
and even in spite of the fragmentary nature of our knowledge of the 
Gordiodrilid groups, as good as, if not better than those in some other 
portions of the family. Thus for instance Kerriona is distinguished 
from Eukerria, aside from the larger size and pigmentation, only by a 
wider pairing of the setae in a portion of the axis, all of which are char- 
acteristics at times of specific value only or none at all. Haplodrilus 
is distinguished from Ocnerodrilus only by its proandry or from Euker- 
ria merely by its microscolecine male terminalia. Py-maeodrilus is 
distinguished from the African section of the subgenus Ilyogenia only 
by diverticulation of the spermathecae (a characteristic apparently 
not of generic value even in a restricted Gordiodrilus) and the approxi- 
mation (rather than actual union) of male and prostatic pores, a dis- 
tinction not considered of generic value in Malaharia or certain non- 
Ocnerodrilid genera. Admittedly some of the characteristics sug- 

1 In the Megascolecid Hoplochaetella the spermathecal pores are ahnost universally released 
from an intersegmental location. 



gates: peregrine earthworms 85 

gested for generic definition are not of the usual sort but the taxonom- 
ist must be concerned first of all with the actual morphology of his 
animals rather than the vague ghosts of idealized forms born from 
phylogenetic speculations. 

Of the Ocernodrilid fauna of the African continent certainly but a 
very small fraction is now known, and almost as soon as new material 
becomes available for study new genera will be necessary. The infor- 
mation available even now indicates that evolution within the family 
has not been limited to the very restricted series of developments that 
have been considered hitherto in Megadrilid phylogeny. Whether or 
not any of the new genera necessitated by more extensive collecting 
will be definable as suggested above of course remains to be seen. In 
view of the difficulties associated with the small size of animals one to 
two mm. in diameter, inability to reexamine old material, lack of 
newer material from critical localities, the inadequacy and contradict- 
tory nature of so much of the information that is available, and the 
limited series that have been studied in the past,"- a satisfactory revi- 
sion of Gordiodrilus is impossible at present. Accordingly nothing that 
has been said above should be taken as a designation of a genotype for 
Gordiodrilus or as in any way restricting the genus. Until such time 
as revision becomes practicable, the various groups may be referred 
to as Nannodrilid, Diaphodrilid or by species names, as "a robustus 
group." 

Gordiodrilus peguanus sp. nov. 

Material examined. From Burmese collections: 

Moulmein, Amherst district, October, 3 aclitellate and 42 partially clitellate 
specimens. K. John. 

Mupun, Amherst district, October, 6 partially clitellate specimens. K. 
John. 

Boyag>a, Thaton district, October, 31 partially cUtellate or clitellate speci- 
mens. K. John. 

Taungzun, Thaton district, August, 1 clitellate specimen. K. John. 

Kyaikto, Thaton district, August, 1 clitellate specimen; September, 3 
cUtellate specimens. K. John. 

Sittang, Thaton district, October, 1 juvenile, 53 partially clitellate and 2 
clitellate specimens. K. John. 

Kyauktan, Hanthawaddy district, September, 7 clitellate specimens. 
K. John. 

Syriam, Hanthawaddy district, September, 1 clitellate specimen. K. John. 

' Twenty eight species known only from types or original descriptions, seven known only 
from unique types, seven known only from short series of two (3 species), three (3) or four (1) 
specimens. Much, if not all of the material available has doubtless been of a disappointingly 
refractory nature for investigation by means of sections. 



86 bulletin: museum of comparative zoology 

Kungj^angon, Hanthawaddy district, September, 11 clitellate specimens. 

K. John. 
Rangoon, Hanthawaddy district, June, 33 clitellate specimens; February, 

21 clitellate specimens; March, 18 clitellate specimens. K. John. 
Hmawbi, Insein district, September, 1 clitellate specimen. K. John. 
Minbu, Alinbu district, September, 5 clitellate specimens. K. John. 
Indaw Lake, Katha district, September, 1 clitellate specimen. Saw San 

Thwe. 

From Indian collections: 

Bangalore, South India, 3 aclitellate and 23 clitellate specimens. Prof. 
C. R. N. Rao. 

Exicrnal characteristics. Length 35-47 mm. Diameter 1-13^ mm. 
Unpigmented. The prostomium is epilobous, tongue short (30). No 
dorsal pores. Nephropores? The setae begin on ii on which all four 
couples are present; behind the clitellum ah and cd about equal, aa 
slightly smaller than he, dd ca. = 3^C. 

The clitellum is dark red, reddish brown, dark yellow, bright yellow 
or (partially clitellate specimens) light yellow or light brown, annular 
except on xvii-xviii, not protuberant, extending from a portion of 
xiii — usually only postsetal — to 18/19 or slightly onto xix, 19/20 
or even onto xx (Bangalore worms), the anterior and posterior boun- 
daries indistinct externally but clearly visible at mid-dorsal incisions; 
intersegmental furrows lacking, setae present. 

The sperm athecal pores are on 7/8 and 8/9, usually with centers on 
or close to h, a very narrow marginal area immediately in front of and 
behind the aperture often slightly tumescent so that the aperture at 
first appears to be large and transversely slit-like. Rarely a single 
pore may be in the median half of he. Female pores are on or just 
lateral to h, only slightly nearer to the setal arc of xiv than of xiii, 
hence close to site of 13/14. 

On xvii and xviii of a fairly large proportion of the specimens there 
is a white, transversely placed, slightly raised, almost square male 
shield, extending anteroposteriorly to or almost to the sites of 16/17 
and 18/19 and laterally into the median portion of he, usually fairly 
level, the boundaries of the shield indicated only by elevation and 
epidermal whitening. The surface of shield is not as smooth as the 
epidermis of the surrounding area. The seminal grooves are practically 
straight, though margins of the grooves are slightly irregular, and ex- 
tend anteroposteriorly between the setal arcs of xvii and xviii, in the 
region of ah. At the anterior and posterior ends of the grooves the 
margins may be very slightly swollen. Ventral setae of xvii and xviii 



gates: peregrine earthworms 87 

may be lacking (invisible only?) or present in part and displaced 
mesially; a (presumably) of xvii present on both sides (6), a and h of 
xvii present on both sides (1), a (presumably) of xvii and xviii present 
on both sides (30). Prostatic and male pores have not been identified 
definitely but are probably represented by minute, greyish translucent 
spots at the termini of the grooves (prostatic pores) and midway be- 
tween the termini (male pores). 

On most of the other specimens the unpaired male field is replaced 
by a pair of longitudinally dumb-bell-shaped, whitened, areas with 
sharply delimited margins, the wider rounded ends of the porophores 
much more protuberant than the middle portions, k midventral 
region between the two areas is usually slightly depressed longi- 
tudinally and in addition there may be slight transverse depressions 
just anterior and just posterior to the areas. The midventral region 
between the porophores may or may not be colored like the clitellum. 
On the few remaining specimens the condition of the male field appears 
to be intermediate between the extremes just described. 

Genital markings are lacking. On one worm seta h of xvi, on each 
side, is in a tiny, whitened, transversely placed tumescence. 

Internal anatomy. All septa from 5/6 posteriorly are present; 5/6 
is membranous and delicate, 6/7-9/10 slightly stronger but almost 
transparent. 

There is no gizzard but the oesophagus in v may be slightly widened 
in such a way as to look somewhat like a very rudimentary gizzard. 
Anterior to 5/6 on each side there is a large mass of glandular(?) 
tissue, characterized in part at least by a brilliant iridescence, the 
masses covering over the gut dorsally. Similar but smaller glands are 
present in \d-viii on the anterior faces of the septa just lateral to the 
gut and median to commissures and nephridia, the size decreasing 
posteriorly. On the posterior face of 8/9, just lateral to the gut on 
each side, there is a tiny disc also characterized by a brilliant iri- 
descence. The calciferous sac is large, rather squarish, dislocating 9/10 
posteriorly, opening into the gut through a transversely or longi- 
tudinally slit-like aperture in the floor of the oesophagus posteriorly 
in ix, the lumen of the sac small, transversely elliptical to triangular in 
cross section, nearer the ventral surface of the sac in the anterior por- 
tion. On the inner wall of the oesophagus in \-viii there are low, 
longitudinal, somewhat irregular whitish ridges. Midsegmentally or 
posteriorly in ix the oesophagus is abruptly widened and remains wide 
through x and xi. The intestine begins in xii (30), the oesophageal 
valve small and anteriorly in xii. No typhlosole (4). 



8b bulletin: museum of comparative zoology 

The dorsal blood vessel (single) is continued anteriorly to the region 
of the cerebral ganglia. A supra-oesophageal vessel is visible on the 
gut in the posterior portion of ix where it receives on each side six to 
eight parallel vessels from the lateral wall of the oesophagus. Posterior 
to 9/10 the vessel is usually unrecognizable, but in several specimens 
a partly empty continuation can be traced to a posterior portion of xi 
where it bifurcates, the branches passing ventrally on the lateral faces 
of the gut. Extra-oesophageal trunks are recognizable in v-viii just 
ventrolateral to the gut, in ix passing onto the ventral face of the 
calciferous sac where they disappear. No subneural. The hearts of x 
and xi bifurcate dorsall}', one branch passing to the supra-oesophageal 
or its site (in x) or into the bifurcation of the supra-oesophageal trunk 
(in xi), the other branch into the dorsal trunk. The last pair of hearts 
is in xi (30). The commissures of ix are usually as large as the hearts of 
X but open only into the dorsal blood vessel. 

Nephridia from xix posteriorly are large, usually flattened out 
against the parietes and in contact with both septa of a segment, ex- 
tending from a nearly to the mid-dorsal line. The preseptal funnels are 
close to the parietes in ah. The slender ducts pass into the body wall 
at or median to c, in a presetal portion of the segment. Occasionally 
median portions of nephridia lack the yellowish, granular investment 
and are recognizable as tubular loops. Nephridia of xiii-xviii are 
transversely placed loops, each nephridium with very little or almost 
none of the granular material. From xii to vi nephridia are located on 
the anterior faces of the septa and are without investment. Anterior 
to vi nephridia have not been found. 

The male funnels are free in x and xi, usually with a brilliant sperma- 
tozoal iridescence. The testes are vertically placed, anteroposteriorly 
flattened flaps which reach upwards to the level of the gut or rarely 
even to the dorsal parietes, attached to the posterior faces of 9/10 and 
10/11 close to the ventral parietes. The dorsal ends of the largest 
testes are pointed as a result of an accordion-like folding of a thinner 
dorsal portion of the organs. When unfolded tiny patches of iri- 
descence, separated by translucent — almost transparent areas, are 
visible. Deferent ducts are usually recognizable only in xvii or in xvi 
and xvii, in the latter segment just lateral to the ectal end of the pro- 
static ducts, passing into the parietes midway between the prostatic 
ducts of a side. There is a pair of seminal vesicles in xii (30). A struc- 
ture that appears to be a seminal vesicle was found in ix on the right 
side of one specimen, paired seminal vesicles present in ix in three 
Bangalore specimens. Prostates are three to four mm. long, the ducts 



gates: peregrine earthworms 89 

about one mm. long, slenderer than the glands and with a very slight 
(muscular?) sheen, passing into the parietes in xvii and xviii (15), the 
posterior duct nearer to 18/19 than to 17/18. 

The spermathecal duct is longer than the ampulla from which it is 
sharply demarcated, not especially narrowed in the parietes, the wall 
thick and the lumen narrow. A middle portion usually somewhat 
nearer to the ampulla than to the ectal end is slightly thickened so 
that the duct has a rather spindle-shaped appearance. In the thickened 
region there are two or more small chambers; one on each side, two 
on one and one on the other, several on each side, etc., or rarely a com- 
plete circle around the duct. Chambers are always empty. No sperma- 
tozoal iridescence has been noted in any of the spermathecae but in 
absence of evidence to the contrary the chambers may be called 
seminal. Ampullae may contain opaque, white or red material. 

Abnormalities. One specimen from Thaton has its left seminal groove 
on xviii and xix, its left ovary in xiv, left hearts, testes and male 
funnels in xi and xii, left spermathecae in ix and y. The oesophagus in 
X has a thick wall like that of the calciferous gland, especially ven- 
trally. 

Another worm lacks the right prostate of xvii, the left prostatic duct 
of xviii bifurcating near the ental end, one branch passing to a normal 
prostate, the other branch to a short and somewhat irregular gland. 

One Bangalore worm has a well developed clitellum on xvii and 
xviii only. The portion anterior to xvii is probably an abnormal 
regenerate, with more than 16 segments and no calciferous gland. 

Remarks. The unusual size of the structures referred to as testes and 
the presence in the dorsal ends of those structures of flecks of iri- 
descence apparently indicates presence in this species of testis sacs 
similar to those of Ocnerodrilus occidentalis. 

Oviducts of Bangalore worms may contain ova in a swollen ental 
portion, seven ova present in one duct, three in another. 

G. peguanus is distinguished from the Burmese unicus by the 
absence of prostates in xix and restriction of seminal grooves to xvii 
and xviii; from the Indian travancorensis by the presence of prostates 
in xvii, the discrete and anteriorly located male pores, and the absence 
of prostates of xix; apparently close to paski, from which it is dis- 
tinguished by the setal ratios (aa<bc rather than aa>bc), slightly 
greater extent of the clitellum (onto xix or xx), epilobous prostomium, 
presence of testis sacs, and presence of nephridia anterior to xii. Just 
how many of these distinctions are actual or taxonomically valid re- 
mains to be determined. 



90 bulletin: museum of comparative zoology 

Diagnosis. Quadriprostatic, pores at termini of straight seminal 
grooves in ab, between the setal arcs of xvii and xviii. Male pores in 
seminal grooves, midway between prostatic pores and about at site of 
17/18. Seminal grooves on paired, longitudinally dumb-bell-shaped 
porophores or on a transversely placed, slightly elevated shield reach- 
ing to or nearly to 16/17 and 18/19 and laterally into the median por- 
tion of be. Setae: ab = cd, aa<bc, dd = 3^2C; a setae of xvii-xviii 
usually present. Quadrithecal, spermathecal pores on 7/8-8/9, on b. 
Clitellum annular, on xiii-xix or xx. Nephropores? Prostomium 
epilobous. Unpigmented. Length 35-47 mm. Diameter 1-13^ mm. 

Holandric, seminal vesicles in xii. Spermathecal duct slightly 
spindle-shaped, longer than the ampulla, with two or more seminal 
chambers in wall of a thickened middle portion. 

Distribution. Amherst, Thaton, Hanthawaddy, Insein, Minbu and 
Katlia districts, Burma. Mysore State, India. 

Genus Malabaria Stephenson 

1924. Malaharia + Aphanascus Stephenson, Rec. Ind. Mas. XXVI, pp. 356 
and 360. (Genotypes M. paludicola and A. oryzivorus Stephenson 
1924.) 

1930. Malaharia -\- Aphanascus, Stephenson, The Ohgochaeta, pp. 857 and 

858.) 

1938. Filodrilus Chen, Cont. Biol. Lab. Sci. Soc. China, (ZooL), XII, p. 422. 
(Genotype F. levis Chen 1938.) 

Diagnosis. Male pores on xvii. Biprostatic, prostatic ducts unite 
with male deferent ducts or open close to male pores. Bithecal, 
spermathecal pores on 8/9. Setae closely paired. Gizzard in vii. 
Calciferous diverticula small, paired, in thickened floor of gut in ix 
and x. 

Distribution.. India (Malabar, 2 species; Travancore, 1 species; 
possibly one species in Mysore), Burma (1 species) and China (1 
species). 

Remarks. Aphanascus is known only from the original account of 
oryzivorus and was distinguished from Malabaria by the absence of 
"prostates" opening to the exterior on xx and by the metandric condi- 
tion of the male organs. The "prostates" in the genotype of Mala- 
baria are associated with genital markings rather than seminal grooves 
and presumably provide no more justification for generic distinction 
than do similar glands and genital markings in Eukerria peguana. 
Confirmation of this \'iewpoint is provided by Aiyer's species, M. 



gates: peregrine earthworms 91 

biprostata, which has only two prostates. Although regarded hitherto 
as of generic significance in the Ocnerodrilidae (except in Gordiodrilus) 
metandry usually is not recognized, in itself, as of sufficient importance 
to warrant generic distinction. In absence therefore of other criteria 
of generic status, retention of Aphanascus, merely because of the 
metandry, seems unnecessary. 

Filodrilns, according to Chen, differs from Malabaria in the absence 
of a posterior pair of "prostates", lack of union of deferent and pros- 
tatic ducts, and presence of an additional pair of testes in ix. Absence 
of posterior "prostates" and of terminal junction of male deferent and 
prostatic ducts characterize M. biprostata and do not appear to be of 
generic significance. (In Microscolex, to mention only one example, 
male and anterior prostatic pores may or may not be united.) Super- 
numerary testes, possibly nonfunctional, even when associated with 
extra male funnels, scarcely warrant generic distinction any more than 
in Pheretima anomala and Lamprodrilus satyriscus Michaelsen 1903. 
Oesophageal diverticula of Filodrilus are so described^ that it is im- 
possible to determine whether or not these structures are the same as 
in Indian species but there is no evidence available to indicate that the 
calciferous region is significantly different from that of Malabaria. 
Unless Filodrilus can be distinguished by important characteristics of 
the calciferous region of the gut there seems to be no justification for 
its retention. 

The distribution of Malabaria is of particular interest in connection 
with the phylogeny of the Ocnerodrilidae. Evolution of the first 
earthworms, the differentiation of the several families and the evolu- 
tion of the numerous genera is restricted by Stephenson to the Tertiary 
and Quaternary. The Ocnerodrilidae, "the last of the stems" derived 
from the root genus of the Megascolecidae, and by the development 
of two pairs of external calciferous sacs in ix and x, is thought to have 
originated in tropical or subtropical Africa from whence it "radiated 
off branches to iVmerica on the west and India on the east" (Stephen- 
son, 1930, p. 835). The eastern branch is composed of Maheina with a 
supposedly endemic species on the Seychelles, Curciia with a supposed 
endemic in South India, and Malabaria with endemics in India and 
Burma. Malabaria with calciferous sacs reduced to vestigial structures 

• Oesophageal diverticula are characterized as follows, "no conspicuous calciferous glands, 
its wall thickened and vascular in viii-x, ventral wall in ix and x specially thicitened with 
rudimentary lumen" (p. 422) and "Calciferous glands in viii— x no special pouches, its wall 
thick and invested with loosely connective tissue and blood capillaries, constricted interseg- 
mentally undoubtedly being rudimentary calciferous glands, portion in ix & x thickest on 
ventral side with rudimentary lumen, sac-like portion about 0.4 mm. in diameter of whole sac, 
about 0.06 mm. in thickness of dorsal wall; thickest ventral wall about 0.17 mm." No figures. 
(Chen, 1938.) 



92 bulletin: museum of comparative zoology 

concealed within the ventral wall of the oesophagus apparently is to be 
considered, so far as that characteristic is concerned, the most highly 
specialized genus of the family. 

A number of questions in connection with these ideas need very 
careful consideration which is not possible here. It should be noted 
however that all of the Ocnerodrilids of the eastern branch are of the 
small size characteristic of so many of the species that have been 
transported across oceans and successfully colonized in new areas. 
Until after rigorous exclusion of the possibility of transportation of 
species of the eastern branch into the areas from which they are now 
alone known, endemicity in the oriental region, eastward radiation 
and any zoogeographical conclusions based thereon may be regarded 
as under suspicion. Arguments for evolution in a portion of the 
eastern branch by regression of the calciferous sacs to intra-oesophageal 
rudiments are not very convincing and a possibility that the intra- 
oesophageal diverticula of Malabaria are primitive or even without 
genetic organ relationship to external oesophageal sacs requires 
examination. 

Key to the species of Malabaria 

1. a. Holandric 2 

b. Metandric oryzivora (Stephenson) 1924. 

2. a. Prostate-like genital marking glands present, with apertures 

on XX paludicola Stephenson 1924. 

b. Prostate-like genital marking glands lacking 3 

3. a. Spermathecal ampulla irregularly saccular and longer than the 

duct biprostata Aiyer 1929. 

b. Spermathecal ampulla shortly ellipsoidal and much shorter 
than the hypertrophied duct levis (Chen) 1938. 

A Mysore form, possibly specifically distinct, differs from biprostata 
and levis in the location of spermathecal pores in mid be rather than 
onb. • 

Malabaria levis (Chen) ? 

1938. Filudrilus levis Chen, Cont. Biol. Lab. Sci. Soc. Zool. XII, p. 423. (Type 

locality Fo-peng, Hainan Island. Types?) 
Material examined. From Burmese collections: 

Kungyangon, Hanthawaddy district, September, 1 aclitellate specimen. 

K. John. 
Twante, Hanthawaddy district, September, 2 aclitellate specimens. K. 
John. 



gates: peregrine earthworms 93 

Kayan, Hanthawaddy district, September, 1 aclitellate specimen. K. 

John. 
Pyapon, Pyapon district, September, 16 aclitellate specimens. Maung 

Ohn Maung. 
Kyaiklat, Pyapon district, September, 1 aclitellate specimen. Maung Ohn 

Maung. 
Danubyu, Maubin district, October, 2 aclitellate specimens. Maung Ohn 

Maung. 

External characteristics. Length 40-87 mm. Diameter ca. 1 mm. 
Unpigmented. Prostomium epilobous, rarely proepilobous. Dorsal 
pores lacking. Nephropores unrecognizable (on b ?). The setae begin 
on ii on which all four couples are present; ab and cd about equal, aa 
and be about equal, dd ca = 3^C. No epidermal modification recog- 
nizable in clitellar region at mid-dorsal incisions. 

Spermathecal pores are small transverse slits on 8/9, on or just 
lateral to b. Female pores have not been identified but the sites are 
probably indicated by slight tumescences on or just lateral to b and 
just behind 13/14. 

The male terminalia are variable. Nine specimens have paired 
prostatic pores, five specimens have no prostatic pores while the other 
worms have a single prostatic aperture on the right or the left side. 
Prostatic porophores are located on xvii in the region of ab with 
centers about on b, one or (usually) both ventral setae lacking when 
the porophore is present, setae present and normal when the poro- 
phore is lacking. On one specimen each porophore is a tiny, trans- 
versely placed area of shortly elliptical outline with a clearly demar- 
cated but narrow rim central to which the surface is slightly depressed 
and pitted or roughened. Pores are unrecognizable on this marking 
which is located on an area of slight epidermal thickening that reaches 
anteroposteriorly towards 16/17 and 17/18. On other specimens a 
single minute pore may be recognizable in a slight, transversely placed, 
slit-like depression, the margins slightly tumescent to form anterior 
and posterior lips. The depression may be an early stage in develop- 
ment of a parietal invagination such as characterizes Eukerria peguana. 
On Pyapon worms prostatic porophores are rudimentary, even at best 
development, usually only a tiny conical protuberance on line b 
visible, pores unrecognizable. 

Genital markings, usually present, are variable in position as shown 
in the table, and are often rudimentary like the prostatic porophores. 
At best development the marking has a smooth, flat, opaque, whitened 
rim and a sharply demarcated, slightly translucent, central protuber- 



94 bulletin: museum of comparative zoology 

ance of shortly and transversely ellipsoidal appearance. The pro- 
tuberance is due to the presence of a transversely placed, clear vesicle 
which bulges the muscular layers of the body wall rather conspicuously 
into the coelomic cavity, the vesicle containing a whitish coagulum. 

Internal anatomy. All septa from 5/6 posteriorly are present; 
6/7-8/9 thickened and probably with muscular fibres, 9/10 slightly 
thickened. 

The gizzard, in vii (20), is elongate and with marked muscular 
sheen. The oesophagus in viii-x is widened, slightly constricted 
septally, heavily vascularized, with numerous vertical vessels clearly 
visible in the lateral walls. In ix the wall is thickened, especially so 
ventrally. Calciferous sacs are lacking, intramural diverticula un- 
recognizable in dissections. Posteriorly in ix at least, and possibly 
also in x, there is a very slight median groove on the floor of the gut. 
The oesophagus in xi-xii is white and without especial vascularity. 
The intestine begins in xii (20), the oesophageal valve anteriorly in 
the same segment. No typhlosole. 

The dorsal blood vessel (single) is continued anteriorly to the 
region of the supra-oesophageal ganglia. A supra-oesophageal trunk, 
as large as the dorsal blood vessel, terminates posteriorly and abruptly 
in xi at the junction with anterior branches of the hearts of xi, and is 
unrecognizable anterior to 9/10. Extra-oesophageals are recognizable 
only in viii-xi where they are located on the ventral face of the gut 
close to the midventral line. On the parietes, in he, on each side 
(1 specimen), there is in xi-xvi a longitudinal vessel. No subneural. 
The last hearts are in xi (20). Hearts of x and xi bifurcate dorsally, 
the anterior branch large, filled with blood and passing into the supra- 
oesophageal, the posterior bifurcation empty, white, passing to the 
dorsal trunk. 

Nephridia are large, reaching into contact with both septa of a seg- 
ment and upwards to the dorsal trunk, filling the coelomic cavities be- 
tween the gut and body wall. A translucent, slender duct passes into 
the parietes just behind the septa, on or close to h. The neck is even 
more slender and transparent, the preseptal funnel small and almost 
transparent, close to the ventral parietes, about at a. Anterior to xii 
or xiii nephridia have not been found and if present must be much 
smaller. 

The testes are delicate, fan-shaped bodies in the usual positions on 
9/10 and 10/11. Paired structures resembling testes are present on 
the posterior face of 8/9 in a number if not all of the specimens. Male 
funnels are free, delicate and with no spermatozoal iridescence, at the 



gates: peregrine earthworms 95 

usual positions in x-xi. In addition there is, in a number of the worms, 
a pair of funnel-Hke structures on the anterior face of 9/10. Paired, 
lobed, very soft seminal vesicles are present in xi and xii (20). Vasa 
deferentia were not traced but in the Kungyangon specimen a slightly 
iridescent thread in xvi, probably a posterior portion of the left 
deferent duct, disappears from sight within a slight protuberance from 
the parietes into the coelomic cavity close to the ectal end of the 
prostatic duct. Prostates are nearly circular in section, flecked with 
whitish spots. A duct is distinguishable only by a slight translucence 
and absence of whitish flecks. Prostates may extend posteriorly into 
xxviii though long enough to reach several segments further if straight- 
ened out or they may be closely coiled and reaching anteriorly into the 
region of xiv-xv. In one worm the single prostate passes anteriorly 
into XV then posteriorly into xx and then anteriorly again. There is 
always a prostate for each porophore, even that on xviii. 

Ovaries are present in xii in several specimens in addition to the 
usual pair in xiii. 

The spermathecal ducts, in the Kungyangon specimen, are of about 
the same thickness as the prostates, with thin transparent walls except 
in a short portion close to the parietes, looped irregularly, slightly and 
irregularly constricted in part to produce a somewhat moniliform ap- 
pearance, slightly widened at the ental end into a very small, spheroidal 
to shortly ellipsoidal ampulla only about half again as wide as the duct 
but sharply marked off and also transparent. The right spermatheca 
reaches into xix but is looped back on itself. The left spermatheca 
penetrates into xxiii. There are no diverticula or seminal chambers. 
In other worms the spermathecae are often coiled in ix or in viii. 

The longitudinal musculature is uninterrupted over sites of the 
genital markings. 

Remarks. In spite of careful searches in the most favorable months 
of the year only aclitellate specimens were secured. The largest worms 
are probably in a late juvenile or pre-sexual stage (note complete 
absence of iridescence on male funnels and in spermathecae and the 
juvenile condition of ovaries). The Kungyangon specimen appears to 
be more mature than the others. 

Constant presence of all setae on xviii-xix and absence of epidermal 
modifications on those segments show that the male pores are not on 
xviii and xix and the disappearance of the male deferent duct into the 
parietes along with the prostatic duct indicates a location on xvii. It 
was impossible, either from dissections or sections, to determine whether 
deferent and prostatic ducts unite before opening to the exterior. 



96 bulletin: museum of comparative zoology 

In Pheretima anomala Michaelsen 1907 abnormalities such as 
presence of supernumerary gonads and genital funnels as well as 
absence of male genital terminalia" and of spermathecae have been 
attributed to parasitic action and it is therefore of interest to note that 
in one of the smallest athecal specimens of Icvis coelomic cavities of 
the anterior end are filled with sporozoan parasites. None of these 
parasites were found in other specimens. Absence of spermathecae 
and prostates can scarcely be attributed to immaturity. 

Two specimens were sectioned by professional technicians but pres- 
ervation was unsatisfactory, especially in the portion of the gut belong- 
ing to viii-x, and the material was refractory. In sections through the 
calciferous region of the gut a definite median groove in the thickened 
floor of the oesophagus is quite obvious but intramural diverticula 
or non-vascular spaces lined by epithelium are quite unrecognizable. 
Though intramural diverticula, supposedly characteristic, of Mala- 
baria, have not been found the bithecal and biprostatic condition is 
sufficient justification for the generic identification, at least for the 
present. 

The Burmese worms described above are clearly distinguished from 
the three Indian species of Malabaria by the hypertrophied spermathe- 
cae. Distinction from Chen's lems is doubtful. Assuming that the 
calciferous region of the gut is similar in both, the Burmese worms ap- 
parently differ only in a very slightly more lateral location of the 
spermathecal pores and the location of male and prostatic pores in 
parietal invaginations (in levis male and prostatic pores appear to be dis- 
crete and superficial, the male pore just median to the prostatic pore, 
while spermathecal pores are in ah). Against these rather unimportant 
and possibly fictitious difl^erences are to be considered the similarities 
of such unusual sorts as: hypertrophied spermathecal ducts, presence 
of testes and male funnels in ix and of ovaries in xii, frequent absence of 
male genital terminalia and spermathecae. 

A satisfactory diagnosis cannot be given until fully mature normal 
specimens with properly preserved calciferous region have been studied. 

Diagnosis. Male and prostatic pores in parietal invaginations 
opening to exterior by transversely slit-like apertures with centers on 
6? Genital markings on 8/9 or viii, rarely on xviii, each marking con- 
taining a clear gland. Spermathecal pores on or just lateral to b. 
Setae: ab = cd, aa=bc, dd=}/^C; ventral setae of xvii lacking. Clitel- 
lum? Nephropores on b? Prostomium epilobous. Unpigmented. 
Length to 87 mm. Diameter 1 mm. 

Holandric; seminal vesicles in xi and xii; an extra pair of testes and 



gates: peregrine earthworms 



97 



male funnels (occasionally only?) present in ix. Male deferent ducts? 
Spermathecae long enough to reach posteriorly into region of xxv, 
duct sharply marked off from and several times longer than the 
spheroidal to shortly ellipsoidal ampulla. 

Distribution. Hanthawaddy, Pyapon and Maubin districts, Burma. 
Hainan Island, China. 



Variation in Burmese specimens of M. levis 











6 
g 










.2 o^.Sf 


_i 




•T. 




B 

3 


68 


2 c II II 


OS 


E 

K 


a 
i 




1 


1 


viii, lef t side just be- 


Right to xix 


Genital marking 


Kungyangon 








hind ab. 


left to xxiii 


with wide rim. 




2 


40 


r&l 
right xviii 





Coiled in ix 


Ovaries in xii 


Twante 


3 


52 


r 


viii, right side just 
behind ab 


Coiled in ix 






4 


85 


r &1 


xviii, just in front 
of ab 


Into viii 




Kayan 


5 


87 


1 


viii, left side, just 
median to a and 
just in front of 8/9 







Pyapon 


6 


65 





8/9, median to a, 
left side 









7 


70 





8/9, center just 
median to a 









8 


69 


1 


8/9, median 





Testes and male 
funnels in ix 




9 


58 


r&l 





Coiled in ix 






10 


60 





viii, left side, just 
behind ab 





Testes and male 
funnels in ix, 
ovaries in xii 




11 


63 





8/9, leftside, center 
on a 









12 


52 


r&l 


viii, left side, just 
behind ab 


Into xxv 


Prostates into 
xxxviii 




13 


85 





8/9, right side 







'* 


14 


48 


r&l 





Coiled in ix 




" 


15 


35 


r&l 


8/9, left side, just 
median to a 


Coiled in ix 






16 


31 


r&l 





Into viii, then 
to XV 






17 


54 


r 


8/9, right side, just 
median to a 


Coiled in ix 


Genital marking 
with wide rim 




18 


55 


r 


viii, right side just 
behind ab 


Into xix 


Genital marking 
with wide rim 




19 


41 


1 


8/9, left side, just 
median to a; xviii 
left side just in 
front of ab 





Testes and male 
funnels in ix 




20 


34 


1 


8/9, left side, just 
median to a 


Into xxi 


Ovaries in xii. 
Genital mark- 
ing with wide 
rim 




21 


40 


r&l 


8/9, right side, just 
median to a 


Into xi 




Kyaiklat 


22 


38 


r&l 





Coiled in ix 




Danubyu 


23 


49 


1 


8/9, left side, center 
just median to a 


Into xxi 


Prostate 5 mm. 
long 


.. 



98 bulletin: museum of comparative zoology 

Ma LAB ARIA species 

Material examined. From Indian collections: 

Bangalore, South India, 1 juvenile and 3 aclitellate specimens. Prof. 
C. R. N. Rao. 

External characteristics. Length to 61 mm. Diameter to Ij^ mm. 
Unpigmented. Dorsal pores lacking. Setae begin on ii ; on xx, a6 = C(f, 
aa > be, ventral setae of xvii lacking. 

Spermathecal pores are on 8/9 with centres at mid be. Female pores 
probably are on 6 close to 13/14. Male porophores are transversely 
placed, rather translucent, slightly protuberant areas on xvii with cen- 
tres on or close to b, the appearance suggestive of a clear gland in a 
parietal invagination but with the latter practically in the initial stage 
of development. Each protuberance is at the center of a circular area 
of epidermal whitening that extends well into be and aa. On one speci- 
men with especially marked protuberances there is on each side a 
transversely slit-like invagination in which clear glands apparently 
are lacking. 

A single genital marking like a clear vesicle of levis, presetal in ab, 
is present on the left side of ix of the juvenile specimen. 

Internal anatomy. A rudimentary gizzard is present in vii. The 
intestine begins in xii. No typhlosole. The floor of the gut in ix is 
markedly thickened. In x the gut wall is slightly thickened but not 
especially so ventrally. Intramural diverticula if present were un- 
recognized. 

A supra-oesophageal trunk is present in ix-xi. Extra-oesophageals 
pass onto the ventral surface of the gut in ix. Hearts of x-xi are latero- 
oesophageal. 

Nepliridia are large from xiv posteriorly, small in xiii, unrecognizable 
anteriorly. 

Male funnels are present in x and xi and seminal vesicles in xii. Pro- 
states extend into lii but are long enough to reach much further. The 
vas deferens disappears from sight in a circular patch of parietal thick- 
ening into the center of which the prostatic duct passes. In the juvenile 
specimen there are no prostates or parietal modifications in xvii and 
the male deferent duct apparently passes into xxiii where it disappears 
from sight just behind the ventral setae. In this specimen all setae of 
xvii-xix are present. 

The spermathecal duct is about as long as the ampulla, the ectal 
half thickened and with slight muscular sheen. The ampulla is of 
shortly elliptical outline, flattened and clearly marked off from the 
duct. 



gatp:s: peregrine earthworms 99 

Remarks. The worms were so soft that the\- broke apart during 
study of external characteristics or in course of dissection, thus pre- 
venting more adequate characterization. Assuming that the structure 
of the gut in x-xi is such as to warrant generic identification as Mala- 
baria (there is no other bithecal genus without calciferous sacs), the 
more lateral location of the spermathecal pores appears to indicate 
specific distinctness from forms already named. 

The continuation of the male deferent ducts posterior to the male 
pore segment as if in search of nonexistent prostates, in the juvenile 
specimen, is an exact parallel of a condition frequently characteristic 
of aprostatic individuals of Pheretima anomala. 

Genus Ocnerodrilus Eisen 1878 

OCNERODRILUS OCCIDENTALIS Eisen 

1878. Ocnerodrilus occidentalis Eisen, N. Acta Soc. Upsala, (3), X, (4), p. 10- 

(Type locality Fresno County, California. Types?) 
Material examined. From Burmese collections: 

Rangoon, Hanthawaddy district, September, 3 clitellate specimens. 

K. John. 
Kaj^an, Hanthawaddy district, September, 1 clitellate specimen. K. John. 
Taik-kyi, Insein district, September, 1 clitellate specimen. K. John. 
Wanetchaimg, Insein district, September, 1 clitellate specimen. K. John. 
Toungoo, Toungoo district, October, 1 clitellate specimen. K. John. 



Rangoon, Hanthawaddy district, June, 1 clitellate specimen. K. John. 

External characteristics. Length 20-23 mm. Diameter 1 mm. Unpig- 
mented. Prostomium? (Buccal cavity everted and first two segments 
softened.) Dorsal pores lacking. Nephropores? The setae begin on ii 
on which all four couples are present; on segments just behind the 
clitellum ah and cd are about equal, aa<bc. The clitellum is dark red- 
dish, annular, not protuberant, boundaries not clearly demarcated but 
probably extending from a postsetal portion of xiii to 20/21; interseg- 
mental furrows lacking, setae present. There are no spermathecal 
pores. The female pores are transversely placed slits on xiv, on b, 
probably about midway between the setal arc and site of 13/14. 

The male pores are on xvii at the ventral ends of tiny, conical pro- 
tuberances from the central portions of nearly circular, fairly sharply 
demarcated porophores extending from a lateral portion of aa into the 
median portion of be, the male pores just lateral to b. Just median to 
the male pore on the right side of one specimen there are visible the 



100 bulletin: museum of comparative zoology 

tips of two setae, a and b, but on the left side of the same worm only 
one seta is visible. 

Internal anatomy. A vertically placed glandular mass on each side 
of the oesophagus in vii and viii is characterized by an iridescent ap- 
pearance. Between 6/7 and the posterior end of the pharyngeal bulb 
there are further iridescent masses. In vi-vii the gut is widened and 
with stronger wall but no especial muscularity is recognizable. The 
oesophagus in ix is dorsoventrally flattened and rather broad. Into the 
lateral face of this portion of the gut there passes on each side the very 
short stalk of a rather pear-shaped, relatively large calciferous sac 
which hangs ventrally beneath the oesophagus. The intestine begins 
in xii. No typhlosole. 

The dorsal blood vessel (single) is continued anteriorly into the region 
of iii. A supra-oesophageal trunk is present in ix-xi. The hearts of 
x-xi bifurcate dorsally, the anterior branch passing onto the gut, the 
posterior branch to the dorsal trunk. No subneural. 

Nephridia of the Rangoon worms are flattened out against the body 
wall, in contact with both septa of a segment, extending from h well 
towards the mid-dorsal line. A portion of each nephridiuin has a 
yellowish, translucent and granular appearance. From xi anteriorly 
the yellowish investment is lacking and from xiv anteriorly the ne- 
phridia are smaller though still flattened out against the body wall. 
Nephridia of the Kayan worm are small, slender loops, not flattened, 
not in contact with either septum of a segment, and with no or almost 
no granular investment. 

The male funnels, two pairs, are free, each funnel with a brilliant 
orange-red iridescence. The testis sacs, in x and xi, are vertical struc- 
tures reaching upwards into contact with the dorsal blood vessel. 
There are no seminal vesicles. The elongately tubular prostates, circu- 
lar in cross section, extend back into xxv or xxvi but are long enough 
to reach into xxviii-xxx when uncoiled. A duct is scarcely recogniza- 
ble. The vas deferens on each side passes into the parietes just anterior 
to the ectal portion of the prostate, union of deferent and prostatic 
ducts within the parietes. No setal follicles are visible in the neighbor- 
hood of the ectal portion of the prostate nor splits in the longitudinal 
musculature such as might indicate the presence of setae. 

Ovaries are large but with few though large ova. Oviducal funnels 
are rather large and have thick rims. 

Remarks. The June specimen from Rangoon dift'ers from the other 
worms as follows: clitellum light yellowish, male pores minute aper- 
tures of about the same size as openings into setal follicles, on xvii, 



gates: peregrine earthworms 101 

very slightly posterolateral to b, no porophores, no modification of 
epidermis around the male pores, setae a and b of xvii present on both 
sides; prostates short, nearly straight, transversely placed, smaller 
than the nephridia. The worm appears to be only a slightly abnormal 
specimen of ocddentalis but the shortness of the prostates is suggestive 
of Eisen's var. arizonae. 

Distribution. Probably widely distributed throughout tropical and 
subtropical regions, already recorded from North America, West 
Indies, Africa, China, India and Burma (Hanthawaddy, Insein and 
Toungoo districts). 

Genus Thatonia gen. nov. 

Diagnosis. Male pores on xviii. Quadriprostatic, prostatic pores on 
xvii and xix, at termini of seminal grooves. Quadrithecal, spermathecal 
pores on 7/8 and 8/9. Setae closely paired. Gizzard in vii. Holandric; 
seminal vesicles in xi and xii. 

Genotype. Thatonia gracilis spec. nov. 

Distribution. Burma. 

Remarks. Close to Malabaria from which it is distinguished at 
present by the acanthodriline male terminalia and- the presence of 
spermathecae opening to the exterior in 7/8. Distinction from Mala- 
baria may be questioned. In view of the apparent uniformity of the 
microscolecine male terminalia and the bithecal condition in Mala- 
baria, as well as the apparent validity of such distinctions for generic 
definition elsewhere in the Ocnerodrilidae and also in other families, 
generic status seems preferable to inclusion in Malabaria. 

According to orthodox oligochaete phytogeny, Thatonia with a 
quadrithecal battery, acanthodriline male terminalia and single gizzard 
is primitive but with regard to the calciferous portion of the oeso- 
phagus is highly specialized. Comment on the last assumption has 
been made in connection with Malabaria. A possibility that the 
primitive ancestral Ocnerodrilid had no gizzard may well be worthy of 
consideration. 

Thatonia gracilis spec. nov. 

Material examined. From Burmese collections : 

Taungzun, Thaton district, September, 1 clitellate specimen. K. John. 
Kyaikto, Thaton district, October, 3 aclitellate and 1 cUtellate specimens 
in poor condition. K. .John. 



102 bulletin: museum of comparative zoology 

Thongwa, Hanthawaddy district, September, 21 juvenile and 5 aclitellate 
specimens. K. John. 

Kayan, Hanthawaddy district, September, 4 acHtellate specimens. K. 
John. 

Kyaiklat, Pyapon district, September, 8 aclitellate specimens. Maung 
Ohn Maung. 

Maubin, Maubin district, September, 2 juvenile specimens. Maung Ohn 
Maung. 

Danubyu, Maubin district, October, 6 partially clitellate specimens. 
Maung Ohn Maung. 

Ingabu, Henzada district, October, 1 aclitellate specimen. Maung Ohn 
Maung. 

Zalun, Henzada district, October, 2 aclitellate specimens. Maung Ohn 
Maung. 

Henzada, Henzada district, October, 17 partially clitellate specimens. 
Maung Ohn Maung. 

External characteristics. Length 63-87 mm. Diameter ca. 1 mm. 
Unpigmented. Prostomiura proepilobous or epilobous. No dorsal 
pores. Nephropores have not been identified definitely but may be on 
or near b at the anterior margins of the segments. The setae begin on 
ii on which all four couples are present; behind the clitellum ab and 
cd are about equal, aa a trifle larger than be, dd co. = 3^C; ventral setae 
of ii, iii or iv to ix, x, xi or xii enlarged, especially so on viii or viii and 
ix, lateral setae of the preclitellar segments also somewhat larger than 
on the postclitellar segments but not as large as those of the ventral 
couples. 

The clitellum is light yellowish, saddle-shaped, not protuberant, 
reaching ventrally to b, extending from 12/13, a posterior portion of 
xiii or 13/14 to 22/23, onto xxiii or even to 23/24, anterior and pos- 
terior limits practically unrecognizable externally but clearly visible 
at the mid-dorsal incision; intersegmental furrows lacking, setae 
present. The ventral margin of the clitellum may be unrecognizable 
or practically so or the non-clitellar mid-ventral region may be dis- 
tinguished by a white appearance which ends abruptly at or near b. 

Actual spermathecal apertures have not been seen on most speci- 
mens but are represented by transversely slit-like depressions on 7/8 
and 8/9 with anterior and posterior margins swollen and lip-like, the 
centers of the depression in ab. On one worm from Danubyu a single 
spermathecal pore is clearly visible as a small, transversely placed slit 



gates: peregrine earthworms 103 

on the ventral face of a markedly protuberant, almost spheroidal 
swelling. The female pores are on or just lateral to b, just behind 13/14. 

The genital shield is longitudinally placed, slighfly raised, whitish, 
extending when completely developed from or just behind 16/17 to or 
nearly to 19/20, widest on xvii and xix where it reaches into be, oc- 
casionally nearly to mid be, narrowest on xviii where it may or may not 
reach to a or b. The shield is fairly sharply demarcated on the most 
mature specimens and presumably is an area of epidermal thickening 
though this is scarcely visible in incisions through the male region, 
except around the prostatic pores where the body wall is certainly 
thickened. On less mature specimens there are two longitudinally 
placed greyish translucent bands extending anteroposteriorly between 
the setal arcs of x\di and xix in ab but bent mesially on xviii. As a re- 
sult of this median bending on xviii the shield appears to have a rather 
H-shaped marking. On more mature specimens the greyish bands 
may be median to a and are depressed, with an appearance of rather 
wide seminal grooves. The prostatic pores are at the termini of the 
grooves, usually just in front (xvii) or just behind (xix) the ends of the 
grooves. The male pores, smaller than the prostatic apertures, are on 
xviii, anteromedian to a, on the median margin of or just median to 
the groove. 

On the youngest juveniles the ventral setae of xvii and xix are 
present. Epidermal modifications in the vicinity of these setae are 
either lacking or so slight as to be unrecognizable with high magnifica- 
tion and brilliant illumination. Yet in two such specimens, on dis- 
section, two pairs of prostates were found, each gland about one half 
mm long. Presumably then some epidermal modification has taken 
place even though unrecognized. On older juveniles the b setae or 
more rarely the a setae alone are lacking on one or both segments, or 
the ventral couples of both segments may be missing. The only 
epidermal modification recognizable on these worms is a slight tumes- 
cence at the sites of the missing setae. On partially clitellate speci- 
mens the setae are probably also lacking though on an occasional 
specimen tiny black dots in the epidermis do look like tips of retracted 
setae. Ventral setae of xviii are usually present, possibly always, but 
are small and at times difficult to identify. 

Internal anatoviy. All septa from 5/6 posteriorly are present; 
5/6-8/9 thickened and with muscular fibres. 

In iv-viii, on each side of the oesophagus there are masses of irides- 
cent glandular (?) material. The gizzard is in vii (10), spheroidal or 
shortly ellipsoidal, much longer and thicker than the portion of the gut 



104 bulletin: museum of comparative zoology 

in viii, usually with only slight muscular sheen but with thick and 
strong wall. In ix and x the gut is markedly moniliform, constricted 
septally, much wider than in xi-xii, the wall thickened and extensively 
vascularized. The floor of the gut in ix may be thicker than the rest 
of the wall or such thickening may be unrecognizable. There are no 
calciferous sacs nor are intramural diverticula recognizable. Occasion- 
ally there is visible on the floor of the gut in ix a slight, longitudinally 
placed groove at the median line, the groove not reaching anteropos- 
teriorly to the segmental limits. On the inner wall of the oesophagus in 
xi there are several thick but low, longitudinal whitish ridges. The 
intestine begins in xii (15), the valve anteriorly in the same segment. 
There is no typhlosole. In the region of the prostates the gut may be 
much narrowed in one or two segments and with a series of transversely 
placed, unpaired ventral caeca just behind the narrowed region, one 
caecum in each segment. These conditions were noted in specimens in 
which the prostates are most compactly coiled — having been unable 
to push posteriorly. Possibly the supposed ventral caeca are merely 
the result of unusually deep ventral constriction of the intestine by the 
septa. 

The dorsal blood vessel (single) is continued anteriorly to the region 
of the supra-pharyngeal ganglia. A supra-oesophageal trunk is present 
in viii-xi, in the posterior portion of xi bifurcated, the branches passing 
ventrally on the lateral faces of the oesophagus. The ventral trunk bi- 
furcates just in front of the anterior margin of the subpharyngeal 
ganglia, the branches passing laterally on the parietes. The extra- 
oesophageal trunk is' first visible as a small vessel on the parietes 
parallel and slightly lateral to the nerve cord, just anterior to 5/6 
receiving a larger vessel from the lateral and dorsal parietes, passing 
upwards on 5/6 and into vi where it is lateral to the gut as in vii. In 
viii the trunk is slightly lower, posteriorly in viii or anteriorly in ix 
passing onto the ventral face of the gut close to the median line whence 
it is continued into xii. In some specimens each trunk apparently ends 
by merely disappearing from sight in the posterior part of the segment, 
in other worms the trunk passes off from the gut in an anterior portion 
of xii (with no visible continuation posteriorly on the gut) and to the 
parietes where it runs posteriorly for a short distance parallel and 
slightly lateral to the nerve cord. No subneural. The hearts of x and 
xi bifurcate dorsally, one branch passing to the dorsal vessel, the other 
to the supra-oesophageal (4), both branches empty, white and slender. 
Commissures of ix can be traced to the dorsal trunk only. The last 
pair of hearts is in xi (15). 



gates: peregrine earthworms 105 

The nephridia from xxii or xxiii posteriorly are large, thick, trans- 
versely placed, almost rectangular, compact discs in which looping is 
scarcely recognizable, in contact with both septa of a segment, extend- 
ing from a to or nearly into contact with the dorsal blood vessel and 
filling all of the space between the body wall and the gut. A very 
slender duct (?) passes into the parietes in a presetal portion of the seg- 
ment at or near to 6. In an occasional segment of each of several speci- 
mens a narrow, neck-like region of a nephridium passes mesially just 
anterior to the duct nearly to the nerve cord and then turns anteriorly 
and penetrates not only through the septum but also through septa 
of the next three or four segments, finally terminating in a bulbous 
swelling just in front of a septum, the swelling single or in two distinct 
lobes, in either case quite funnel-like in appearance. Preservation is 
too poor to warrant further description. In several other specimens 
structures that appear to be funnels with normal relations to the an- 
terior septa of the segments were found. In xv to xxii or xxiii the 
nephridia are transversely placed loops, rather delicate, and with a 
few tiny patches of transparent granular material. In xiv and xiii 
nephridia are still smaller but readily recognizable. In xii-x or ix 
nephridia are much smaller, poorly preserved and often lost in washing 
out masses of coelomic coagulum. From ix or viii anteriorly nephridia 
have not been found. 

The male funnels are free in x and xi, well developed but with no 
spermatozoal iridescence (15). Small disc-like bodies in the usual loca- 
tions on the posterior faces of 9/10 and 10/11 appear to be testes. 
Seminal vesicles of xi reach into contact over the dorsal blood vessel 
and fill the available space in the coelomic cavity of the segment. 
Vesicles of xii, in the most mature worms are much larger, in contact 
above the dorsal trunk and pushing 12/13 and 13/14 back into contact 
with 14/15. The deferent duct passes lateral to the prostatic duct of 
xvii and penetrates into the body wall in xviii anterior to seta a. The 
prostates are long, coiled into a compact mass extending through 
several segments or penetrating posteriorly alongside the nerve cord 
into the region of 1-lx. The margins are somewhat irregular and the sur- 
faces are flecked with whitish spots. The mottled appearance is due 
to a close crowding of granules in certain cells. The glands are one-cell 
thick posteriorly. Ducts are slenderer than the glands and may be 
looped in a rather zigzag fashion, passing into the parietes in xvii and 
xix in ah. A circular region in the body wall into the center of which 
the duct passes is firm but not protuberant. The ectal portion of the 
duct can be pulled out from the parietes, apparently intact. 



106 bulletin: museum of comparative zoology 

The spermathecae are large, coiled, in viii and ix, reaching into 
contact with the dorsal parietes. The ampulla which is elongately 
tubular (length 3+ mm.) has a rather thin and translucent wall. The 
duct is very much shorter than the ampulla and slightly narrower, the 
wall thick and lumen narrow but slightly irregular, the duct slightly 
widened as it passes into the parietes. No diverticula or seminal cham- 
bers. Spermatozoal iridescence has not been observed in any of the 
spermathecae. 

Abnormalities. A number of the worms appear to be abnormal. Of 
these 17 are posterobiprostatic, the anterior prostates lacking, the gen- 
ital shield with no anterior widening and ending at or just in front of the 
site of 17/18, the seminal grooves terminating slightly in front of the 
male pores. On these specimens (three from Kay an, one from Taung- 
zun, three from Thongwa, one from Zalun, one from Ingabu, three 
from Danubyu, and five from Henzada) the ventral setae of xvii are 
present. On three worms one prostate only of xvii is lacking, the ante- 
rior widening of the shield lacking on the side of the missing prostate, 
the seminal groove of that side terminating anteriorly just in front of 
the site of 17/18, the ventral setae of the abnormal side present (left 
prostate lacking in two worms from Danubyu, the right lacking in one 
from Zalun). 

Another worm is much more abnormal, the right half of the genital 
shield about one mm behind the left half, four spermathecae on the 
left side two of which are in one segment. 

Remarks. After two failures a complete series of transverse sections 
(10 u thick) through the calciferous portion of the gut were obtained. 
In spite of the fact that the specimen had been killed and preserved 
without benefit of special fixation for histological purposes the preser- 
vation must be considered good. Vascular spaces in the thickened 
floor of the oesophagus are completely filled with blood. There are no 
vertical clefts in regions between vascular lamellae (as in Malabaria, 
vide fig. 38, pi. 33, Stephenson 1924). In the thickened floor of the gut 
are numerous branching canaliculi, mostly seen in transverse section. 
The thickness of the wall of each canaliculus is about the same as the 
diameter of the lumen, the wall very sharply outlined as if by a definite 
membrane both externally and internally (next to the lumen). Nuclei 
and cell walls are quite unrecognizable. In just one section there is 
visible close to the ventral surface of the gut wall and at both sides of 
a median, vertically placed, blood mass, a pair of spaces similar to 
those marked "x" in Stephenson's fig. 39. In other sections one or 
both of these spaces is lacking or one of the spaces may be replaced by 



gates: peregrine earthworms 107 

two or even three smaller spaces. Each of the larger spaces (canals) 
is also provided with a wall sharply marked oflF externally as well as 
next to the lumen and with no nuclei or cell walls recognizable. These 
canals appear to be formed by junction of the smaller canaliculi. 
None of the canals or canaliculi have been traced into the gut 
lumen. 

In two consecutive sections (and here only), midway between the 
outer and inner surfaces of the floor of the oesophagus there is a pair 
of larger spaces, one on each side of the median plane. The walls of 
these spaces are not as sharply marked off as those of the canals and 
canaliculi but are cellular with about ten fairly large nuclei recognizable 
in each section. The width of this cellular wall is much less than that 
of the lumen which is filled by a fine, webby material. There is no 
median groove in the floor of the gut in the region of the sections con- 
taining the canals or the cell-lined spaces. 

Much thinner sections will be necessary for elucidation of the rela- 
tionships of canaliculi, canals and cell-lined spaces but with the canali- 
culi ramifying throughout the floor of the gut there is nothing par- 
ticularly suggestive of external calciferous sacs reduced to diverticula 
and retracted, as it were, into the floor of the gut. . 

Blood in the vascular spaces of the floor of the gut is stained as else- 
where a bright red but blood in the vascular spaces of the roof of the 
gut is uniformly unstained and of a greenish-translucent appearance 
which is retained clear to the dorsal surface where the red coloration 
again appears. 

Diagnosis. Seminal grooves in or mediian to ab, bent mesially on 
xviii to produce a rather H-shaped figure, on a genital shield extending 
nearly to 16/17 and 19/20. Prostatic pores on setal arcs and at termini 
of seminal grooves, male pores anteromedian to a. Spermathecal pores 
in ab. Clitellum saddle-shaped (?), on xiii-xxii or xxiii. Setae: ab — cd, 
aa>bc, dd=}/2C; ventral setae of xvii and xix lacking, setae of some 
preclitellar segments enlarged, especially a and 6. Prostomium epilob- 
ous. Unpigmented. Length 63-87 mm. Diameter 1 mm. 

Spermathecal ampulla much longer than the duct, tubular, coiled; 
duct slightly widened just prior to entrance into parietes. 

Family MEGASCOLECIDAE 

Genus Woodwardiella Stephenson 1925 

In an earlier paper (Gates, 1938, p. 428) it has been pointed out that 
several Indian species hitherto included in the genus Woodwardiella 



108 bulletin: museum of comparative zoology 

must be transferred to other genera. One such transfer has already 
been made, and the exclusion of others leaves but four species in the 
whole area west of Australia. Of these two may be identical, and all 
may have been exported from Australia. A Ceylonese species, JV. uzeli 
(Michaelsen) 1903, known only from a rather brief description that 
contains at least one error (with regard to the setal ratios), is distin- 
guished from Indian and Burmese forms by a more posterior location 
of the first dorsal pore (?), location of spermathecal pores lateral to b 
and just behind 7/8-8/9, constant presence of post-setal papillae on 
xvii "in den Borstenlinien b", presence of paired female pores, and the 
presence of a typhlosole (location of last hearts unknown). 

Australian species are so variable with respect to characteristics 
that may be of generic importance (such as presence or absence of 
calciferous glands) as to indicate a need for further revision of the re- 
maining and larger portion of the genus. Most species are so in- 
adequately characterized that such revision is inadvisable until further 
studies of internal anatomy have been made. 



Woodwardiella javanica (Michaelsen) 

1910. Woodwardia javanica, Michaelsen, Mitt. Mus. Hamburg, XXVIl, p. 93. 
(Type locality Buitenzorg, Java. Types in the Hamburg Museum.) 

Material examined. From Burmese collections: 

Boj^agj'i, Thaton district, October, 4 aclitellate and 4 clitellate speci- 
mens. K. .John. 

Mupun, Amherst district, October, 1 clitellate anterior fragment. K.John 

External characteristics. Length, 55-75 mm. Diameter, 1-13^ mm. 
Unpigmented. Prostomium proepilobous. Nephropores not seen. 
The first dorsal pore is on 6/7 (6). 

The setae begin on ii ; the setal ratios on xx are, so far as can be de- 
termined, about the same as on pumila, ab<aa<or=bc<cd, dd=OT 
slightly greater than aa, posteriorly dd<aa. Anterior to the clitellum 
d seems to be more dorsal than on pumila. 

The clitellum is usually slightly protuberant, light brownish, an- 
nular, extending from 13/14 or from some portion of xiii to 17/18; in- 
tersegmental furrows and dorsal pores lacking, setae present but 
deeply retracted and scarcely visible. 

Quadrithecal, spermathecal pores minute and superficial, on 7/8-8/9, 
on or immediately lateral to a. 



gates: peregrine earthworms 109 

A single, median female pore is visible just anterior to the setal arc 
of xiv, on two clitellate and two aclitellate specimens. 

The male pores (common apertures of the prostatic duct and penis- 
etal follicles) are at the centers of small, slightly protuberant, rather 
indistinctly demarcated porophores of circular to shortly elliptical 
outline (in the latter case transversely placed), each porophore about 
as wide as ah, the center of a porophore slightly lateral to a to just 
median to h. 

Genital markings apparently are lacking, but markings like those 
of pumila, if present, would be unrecognizable on most of these speci- 
mens. 

Internal anatomy. Septum 5/6 is membranous to slightly muscular; 
6/7-11/12 muscular; 12/13 slightly muscular. 

The gizzard is in v (7), with brilliant muscular sheen. The oesopha- 
gus, behind the gizzard, is slightly constricted by the septa and hence 
somewhat moniliform, the inner wall in vii-xiii or xiv provided with 
low, longitudinal non-lamelliform ridges, these ridges crossed in 
vii-ix or viii-ix by circular furrows to mark off rather low, almost 
conical protuberances. The intestine begins in xviii (7), the part in 
xviii rather conical, with narrow portion anteriorly, full intestinal 
width attained only at 18/19. The valve is in xvii or a posterior 
portion of xvii and the anteriormost part of xviii, and is unusually 
narrow. The intestine may be slightly constricted at 18/19 but does 
not have a valvular structure at the constriction. There is no typh- 
losole (5). 

The dorsal blood vessel (single) is continued anteriorly to the 
region of the cerebral ganglia. A supra-oesophageal is present in vii- 
xiii, disappearing from sight at 6/7 and 13/14, as large as the dorsal 
blood vessel when distended with blood, occasionally empty and then 
unrecognizable. Extra-oesophageals are present, apparently much as 
in pumila, but are always empty in part and then unrecognizable. In 
xiv-xviii, on each side there is usually visible a fairly large, latero- 
parietal trunk which passes mesially just anterior to 13/14 and into 
the extra-oesophageal trunk. In one worm the right lateroparietal 
vessel terminates abruptly in xv, but a large branch from the left 
vessel passes underneath the nerve cord to the right side in xvi and 
then posteriorly into xviii. In one specimen each lateroparietal trunk 
bifurcates just anterior to 13/14, the larger branch passing to the 
extra-oesophageal trunk, the smaller passing up onto the dorsal face 
of the gut. No subneural. The last pair of hearts is in xii (7). The 
hearts of x-xii bifurcate dorsally, one branch passing into the supra- 



110 bulletin: museum of comparative zoology 

oesophageal trunk, the other to the dorsal vessel. Commissures of 
ix may be as large as the hearts. The commissures of ix and the 
hearts of x-xii have been traced to the ventral trunk, the commissures 
of viii-vi not so traced. 

The nephridia from xix posteriorly are flattened out against the 
parietes, in contact or nearly so with both septa, extending from a or 6 
to c. The nephridia of xviii are small and anterior to the prostatic 
duct. In xvii-xiv of the aclitellate specimens the nephridia are also 
flattened (decreasing in size anteriorly) while in the clitellate specimens 
the nephridia are not flattened but look like clusters of micronephridia. 
In xiii-v the nephridia are on the anterior faces of the septa, small in 
xiii-ix, slightly larger in viii-vii, large and like clusters of pharyngeal 
micronephridia in vi-v. In the postprostatic segments the body wall 
is covered with a layer of coelomic coagulum (?), as in pumila. 

The male funnels are free in x and xi. The seminal vesicles are fairly 
large, in contact above the dorsal blood vessel, or reaching to the dorsal 
trunk, paired in xi and xii; each vesicle acinous, the lobes pear-shaped, 
the narrowed central portions rather loosely bound together. Segments 
X and xi are filled with a delicate coagulum in which the anterior 
vesicles are imbedded. The vesicles are soft, especially in xi of clitel- 
late worms, and may be destroyed in removing the coagulum. The 
prostates are confined to xviii, slightly dislocating 17/18 and 18/19, 
or just penetrating into xvii or xix. The duct is 13^-2 mm. long, bent 
into a U-shaped loop or coiled, the ectal half thicker but the whole 
duct with a brilliant muscular sheen. 

On the posterior face of each prostatic duct there are two penisetal 
follicles, the follicles united at the ental ends, diverging ectally to pass 
into the parietes separated by a strand of longitudinal musculature. 
Each follicle contains one functional seta and often a shorter reserve 
seta. The setal shaft is variously curved, occasionally with two or 
three curvatures on a single seta. The tip is flattened and narrowed, 
thin and usually bent over to one side, tapering to a sharp point 
(triangular tip). The margins of the ectal half of the shaft are sinuous, 
the sinuosities alternating on opposite sides of the shaft as if due to a 
groove running in a spiral fashion around the shaft. In each sinuosity 
towards the tip and close against the shaft a rather triangular tooth is 
recognizable with oil immersion objective. After drying or twisting of 
the shaft, teeth may stand out much more conspicuously. Measure- 
ments (in mm. by Miss Chapman) are given below. 



gates: peregrine earthworms 111 

Penial setae 

Width Width Width Thickness 

Seta Length base midshaft tip at side 

a 0.60 0.006 0.005 0.0015 0.001 

b 0.58 0.006 0.005 0.002 0.001 

b(r) 0.32 0.003 0.002 .... 0.001 

a 0.69 0.007 0.005 0.002 0.001 

b 0.63 0.006 0.004 0.002 0.001 

b(r)i 0.41 0.006 0.004 0.002 

a 0.60 0.005 0.004 0.001 

b 0.63 0.006 0.004 0.002 

b(r) 0.15 0.002 0.002 

a 0.55 0.005 0.005 0.002 

a(r) 0.40 0.004 - 0.004 0.001 

b 0.59 0.006 0.004 0.002 

b(r) 0.47 0.007 0.005 0.002 

a 0.63 0.007 0.005 0.003 

a(r) 0.36 0.003 0.003 

b 0.62 0.007 0.006 0.002 

b(r) 0.23 0.003 0.002 

1 Ornamentation continued two thirds way down the shaft, (r) reserve. 

The spermathecae are large, reaching up into contact with the 
dorsal parietes. The duct is much shorter than the ampulla. In the 
ectal half, almost confined to the parietes, the lumen is very narrow 
and with a smooth wall of high epithelium surrounded by a muscular 
(?) layer of about the same thickness. Entally the duct is slightly and 
gradually widened, the lumen enlarged but irregular. This ental por- 
tion of the duct is not marked off externally from the ampulla which 
is also gradually widened entally, the duct distinguished in cleared 
spermathecae by the thicker wall. The elongate and slenderly club- 
shaped diverticulum which passes into the lateral face of the duct just 
above the parietes is about as long as half the combined lengths of duct 
and ampulla. There is no external demarcation into seminal chamber 
or stalk, the latter only slightly narrower than the chamber and 
recognized by the thicker wall and narrow lumen. The chamber is 
about as long as the stalk, the sperm mass in the latter straight or 
slightly looped or even with an appearance of slight spiral coiling 
entally. The diverticulum opens into the widened lumen of the ental 
portion of the duct, immediately above the aperture into the narrowed 
lumen. 



112 bulletin: museum of comparative zoology 

Remarks. The worms are not well preserved, softened just behind 
the clitellum, the clitellar region as well as a short portion just behind 
and just in front roughened. Two of the aclitellate specimens, with 
spermatozoa in the seminal chambers of the spermathecae are probably 
postsexual. 

W. javanica has been known hitherto only from the types, the orig- 
inal description not quite as complete as is now desirable. There is 
however no evidence to indicate that the Burmese worms should be 
distinguished taxonomically from the types. Michaelsen presumably 
missed the delicate anterior vesicles imbedded in the testicular 
coagulum. Seminal vesicles are certainly lacking in ix in the Burmese 
worms. 

W. javanica is close to W. callichaeta, affinis, libferti, and molaeleonis 
(Michaelsen) 1907, being distinguished from these southwest Austra- 
lian species mainly by the presence of a single female pore, and the loca- 
tion of the anterior seminal vesicles in xi rather than ix. Earthworms 
of the x\ustralian region, of the small size of javanica, are doubtless 
very imperfectly known. It is therefore possible that both the Java and 
Burma records indicate importations from some eastern portion of the 
Australasian region. 

Diagnosis. Quadrithecal, spermathecal pores on 7/8-8/9, on a. 
Male pores in ab, on xviii, each pore at center of a small, circular to 
shortly elliptical porophore that is about as wide as ab. Female pore 
median. Seta d dorsal, behind the clitellum close to the mid-dorsal 
line so that (/o?<aa. First dorsal pore on 6/7. Unpigmented. Length 
38-75 mm. Diameter 1-1/3 mm. 

Gizzard in v. Intestine begins in xviii. Hearts latero-oesophageal, 
in x-xii. Holandric; seminal vesicles in xi-xii. Prostatic duct 13^-2 
mm. long, looped or coiled. Spermathecal duct much shorter than the 
ampulla, diverticulum about half the length of main axis, into lateral 
face of duct near the parietes. Penial setae 0.55-0.69 mm. long, 0.005- 
0.007 mm. thick at base, 0.004-0.006 at midshaft, 0.0015-0.003 across 
flattened tip; margins of ectal half of shaft sinous, sinuosities alternate 
and at least ectally containing each a rather triangular tooth. 

Distribution. Boyagyi, Thaton district; Mupun, Amherst district, 
Burma; Buitenzorg, Java. 

WOODWARDIELLA PUMILA Stcplienson 

1931. Woodwardiella pumila, Stephenson, Froc. Zool. Soc. London, 1931, p. 
51. (T}T3e locality Bhamo. Type in the British Museum.) 



gates: peregrine earthworms 113 

Material examined. From Burmese collections: 

Sittang, Thaton district, October, 3 aclitellate specimens. K. John. 
Pegu, September, 2 aclitellate and 44 clitellate specimens. K. John. 
"Jungle", west of Pegu, September, 1 aclitellate and 8 clitellate specimens. 

K. John. 
Rangoon, Hanthawaddy district, June, 32 clitellate specimens; September, 

45 clitellate specimens; February, 1 aclitellate and 8 clitellate specimens; 

March, 27 clitellate specimens. K. John. 
Kungyangon, Hanthawaddy district, September, 12 juvenile or aclitellate 

and 1 clitellate specimens. K. John. 
Kyauktan, Hanthawaddy district, September, 2 aclitellate specimens. 

K. John. 
Thameintaw, Pyapon district, September, 1 clitellate specimen. Maung 

Ohn Maung. 
Pyapon, September, 1 clitellate specimen. Maung Ohn Maung. 
DanubjTi, Maubin district, October, 3 juveniles. Maung Ohn Maung. 
Wakema, Myaungmya district, October, 3 aclitellate and 1 clitellate speci- 
mens. Maung Chn Maung. 
Myaungmya, October, 6 aclitellate specimens. Maung Ohn Maung. 
Bassein, October, 1 juvenile and 2 clitellate specimens. K. John. 
Thinbawgyin, Bassein district, October, 1 aclitellate specimen. K. John. 
Padaukchaung, Bassein district, October, 1 aclitellate specimen. K. John. 
Toungoo, October, 4 juvenile specimens. K. John. 

External characteristics. Length, to 76 mm. Diameter 1 mm. Un- 
pigmented. Prostomium proepilobous, occasionally a single median 
furrow continued from the posterior margin of the prostomium along 
the mid-dorsal line towards l/2. Nephropores were not seen. 

The setae begin on ii on which all four couples are present (10). 
Anterior to the clitellum seta d is mid-lateral or perhaps a trifle on the 
dorsal side, from xvii or xviii posteriorly definitely dorsal. On xx, 
ah<aa< or ca. = bc<cd>dd, dd= or >aa, but only a few segments 
posteriorly dd becomes smaller than aa. 

The first dorsal pore is on 6/7 (18), but on a few other worms the 
rather pore-like marking on 6/7 may be imperforate. 

The clitellum is usually slightly protuberant, reddish to brownish, 
setae present, dorsal pores (except that on 13/14) and intersegmental 
furrows lacking; annular, extending from 17/18 to a presetal portion 
of xiv, 13/14, a postsetal portion of xiii or even slightly onto the pre- 
setal half of xiii. The anterior boundary of the clitellum is often quite 
vague. 

Quadrithecal, spermathecal pores minute and superficial, on 7/8-8/9, 



114 bulletin: museum of comparative zoology 

on b or in ab. The location of the pores is variable from one side to the 
other, from one segment to another and from one specimen to another. 
Location on b is perhaps the most common, just median to b also fre- 
quent, an occasional pore only slightly lateral to a. 

The single female pore is median, immediately in front of the setal 
arc, at the center of a transversely placed, whitish area of shortly 
elliptical outline in aa (32). 

The male pores are in ab, just lateral to a, at or near mid ab, or just 
median to b, at or near the center of small transversely placed poro- 
phores of shortly elliptical (rarely almost circular) outline that are 
about as wide as or a trifle wider than ab and which may extend to 
just median to a or just lateral to b, on xviii. On an occasional specimen 
(12) a penial seta protrudes to the exterior through the pore, the pore 
unrecognizable if the seta is conspicuously protuberant. 

The genital marking is a transversely placed, whitened area of 
elongately elliptical outline on 17/18 in aa. The marking is not pro- 
tuberant, not demarcated except by colour differences, often unrecog- 
nizable. A somewhat similar but less obvious area may be present on 
18/19 but crossed by the intersegment furrow. 

Internal anatomy. Septum 5/6 is present but membranous or slightly 
muscular; 6/7-9/10 muscular; 10/11-11/12 slightly muscular. 

The pharyngeal bulb is unusually short. Between the hinder end of 
the bulb and the gizzard there are masses of iridescent (glandular?) 
tissue which come readily away from the oesophagus. The gizzard is 
in V (15), dislocating 5/6 posteriorly in a funnel-like fashion. On the 
inner wall of the oesophagus in ix-xiii there are a number of closely 
crowded, low, non-lamelliform ridges of a grayish translucent appear- 
ance due to the presence of blood beneath the transparent epithelium. 
The intestine begins in xvi (oesophageal valve anteriorly or posteriorly 
in xvi) or xvii (2 specimens from Sittang) but may be much narrowed 
in xviii or xviii and xix presumably as a result of crowding by the pros- 
tates. There is no typhlosole. 

The dorsal blood vessel (single) is continued anteriorly to the region 
of the cerebral ganglia where it bifurcates, the two branches passing 
ventrally on the pharyngeal bulb to the parietes. A supra-oesophageal 
is present in vii to xiii disappearing just behind 6/7 (or continued into 
vi) and just in front of 13/14. The ventral trunk bifurcates at the 
anterior margin of the suboesophageal ganglia, each branch passing 
to the body wall where it subdivides several times. The extra-oesopha- 
geals are first recognizable as longitudinal vessels just behind the cir- 
cumoesophageal nervous commissures where they are formed by the 



gates: peregrine earthworms 115 

union of two or more vertical vessels from the parietes of i-iii, each 
trunk receiving two connectives from the ventral vessel behind the 
suboesophageal ganglia, passing up to a level just below the gizzard in 
v, with a transverse connective to the opposite trunk in viii, passing in 
ix onto the ventral face of the gut close to the mid ventral line and 
thence posteriorly, breaking up into two or three branches which dis- 
appear in xiv. In xiv-xviii on each side, on the body wall under the 
nephridia, there is a lateroparietal trunk which may be nearly as large 
as the dorsal blood vessel. Immediately anterior to 13/14 this vessel 
passes mesially behind the heart of xiii, and then directly into the 
extra-oesophageal or runs upwards to the level of the dorsal face of the 
gut, then turning ventrally to open into the extra-oesophageal. No 
subneural. The hearts of x-xiii bifurcate dorsally, the anterior branch 
passing into the supra-oesophageal, the posterior branch passing into 
the dorsal trunk. The last pair of hearts is in xiii (15). The commis- 
sures of ix-v connect the dorsal and ventral trunks, the hearts of 
x-xiii also passing into the ventral trunk. The supra-oesophageal 
receives numerous vertical vessels from the lateral faces of the gut in 
ix-xiii but these vessels are largest in xiii. 

The nephi-idia from xiv or xv posteriorly are flattened against the 
body wall, in contact with both septa, extending from a or slightly 
lateral to a to c. From xiii or xiv anteriorly the nephridia are not on 
the parietes but on the anterior faces of the septa, small in xiv or 
xiii-x, larger in ix anteriorly, especially so in v-vi and with the ap- 
pearance of clusters of micronephridial tubules. In the postclitellar 
segments the body wall is clothed with fine, whitish threads, the ap- 
pearance like that of a micronephridial fur, but no tubular structure is 
recognizable in the threads, the whitish material regarded as coelomic 
coagulum. 

The male funnels are free in x and xi (10). Segment x may be filled 
with a loose coagulum. The seminal vesicles of xi and xii may be 
large, filling the coelomic cavities of their segments and reaching up 
into contact with the dorsal blood vessel (6), or medium-sized to 
small, vertically placed bodies on the posterior faces of the septa. The 
prostates are racemose, lobed, compact, much like Pheretima prostates 
in appearance, in xviii (4) or xviii-xix (4). The duct is less than 3^ mm. 
long, straight, slender, usually with a slight muscular sheen that is 
especially noticeable ectally. 

Median to each prostatic duct there are two distinct penisetal fol- 
licles. One of these passes into a cleft in the longitudinal musculature 
on the median face of the prostatic duct, the other passing into the 



116 



bulletin: museum of comparative zoology 



body wall slightly more mesially. A follicle usually contains one penial 
seta; several follicles with two setae, the second presumably a reserve 
seta. The length of the setae varies from 0.28-0.42 mm., the diameter 
at the middle of the shaft from 6-8 micra. The shaft is nearly straight 
or with a slight curvature of the ectal end. The ental end may be 
shortly curved like the handle of a walking stick in which case the shaft 
has a slightly bowed appearance. The tip of the seta tapers gradually 
to a hair-like or filamentous process which may be straight or nearly 
so or bent or curved to one side. Along the ectalmost portion of the 
seta the margins have a slightly sinuous appearance. Under the oil im- 
mersion objective the margins appear to be slightly incised as if mark- 
ing off rather elongately spinelike teeth (spirally arranged?). On two 
setae there is an appearance of very fine, almost hair-like spines, in 
(5-8) circles around the shaft. Usually penial setae are not recogniz- 
able externally and when visible only one protrudes through each 
male pore. 



Seta 

f 
f 
r 
r 
f 
f 
r 
f 
f 
r 
f 
f 
r 



Length 

0.31 
0.28 
0.15 
0.13 
0.38 
0.33 
0.07 
0.33 
0.35 
0.24 
0.42 
0.39 
0.23 



Penial setae 








Width 


Width 




base 


midshaft 


Locality 


0.006 


0.007 


Sittang 


0.005 


0.007 






0.009 


0.008 






0.008 


0.007 






0.006 


0.008 






0.005 


0.007 






0.003 


0.003 






0.007 


'0.006 






0.008 


0.007 






0.007 


0.006 






0.007 


0.008 






0.007 


0.008 






0.006 


0.005 






f functional 






r reserve 









The spermathecal duct is slender and usually slightly shorter than 
the ampulla from which it may or may not be clearly marked off 
externally. In an ental portion of the duct the lumen is fairly wide 
and filled with a dark material like that in the ampulla. Ectal to the 
diverticular junction the lumen appears to be very narrow and the 
duct wall thick but this portion of the duct usually is not satisfactorily 



gates: peregrine earthworms 117 

cleared. The diverticulum usually passes into the lateral face of the 
duct slightly ental to the middle of its length and is not marked off 
externally into stalk and seminal chamber, the latter recognizable 
in the cleared spermatheca by a thinner wall and the shortly ellip- 
soidal to sausage-shaped mass of spermatozoa within, the stalk only 
slightly slenderer than the chamber but with a thicker wall and a 
narrow lumen. In several spermathecae massed spermatozoa are 
continued from the seminal chamber through the stalk into the duct 
and in such spermathecae, when cleared, it is possible to see the di- 
verticular lumen turn entally within the duct wall to open into the 
ental chamber of the duct just above the region where the duct lumen 
is abruptly narrowed. All spermathecae examined (from clitellate 
specimens) are characterized by a spermatozoal iridescence in the 
seminal chamber. 

The ovaries are relatively large. Each ovary examined contains 
10-16 full sized ova and at or near the center several much smaller 
ova. On the posterior face of 13/14 there are small paired sacs. In 
two worms these sacs were empty but in two other specimens the 
sacs contained ova. 

Parasites. On some of the Rangoon specimens, especially on the 
clitellum and around the male pores, there are clusters of rotifers. 
These were sent to the British Museum but have not been further 
identified. 

Abnormality. In a worm from Pyapon the dorsal blood vessel is 
double from the middle of xvii to 13/14, single to the middle of xiii 
and double to 12/13, single henceforward. The left heart of xiii is 
lacking in one of the Sittang specimens. 

Remarks. W. pumila is very close to, if not actually con-specific 
with, W. kayankulamensis Aiyer 1929. The latter can be distinguished, 
according to the author's account, only by characteristics of doubtful 
value; more dorsal location of d setae on preclitellar segments, location 
of the male pore "in line with setae a", presence of "a single penial 
setal sac", and absence of "copulatory papillae". In contrast to his 
usual practice Aiyer failed to state definitely whether there is a single 
female pore or a pair of pores which may perhaps indicate that he did 
not actually see these apertures (types possibly not fully mature as 
there are dorsal pores on the clitellum). Copulatory papillae are cer- 
tainly lacking in pumila but the whitened areas certainly deserve 
recognition as genital markings though not always recognizable, 
possibly due to condition. 

Prof. Aiyer has kindly supplied three specimens of his species for 



118 bulletin: museum of comparative zoology 

study but unfortunately two of these are juvenile while the other is 
aelitellate though nearly sexual (male funnels iridescent but no iri- 
descence in spermathecae) . On these worms female pores and genital 
markings are unrecognizable. Anterior to the clitellum the d setae are 
more dorsal than on pumila. Male porophores are unrecognizable but 
male pores are in ab. Spermathecal diverticula pass to the lateral face 
of the ducts except on the right spermatheca of ix of one specimen. 
Two penisetal follicles are present in each worm on each side, the 
follicles separated ectally by a distinct strand of longitudinal muscula- 
ture. All follicles were removed for examination but several were lost. 
All setae from a follicles except one have truncate tips (worn?) while 
setae from b follicles have tips terminating in a spine much like that of 
pumila. Ectal portions of a setae at least are more strongly curved 
than in pumila. The exceptional a seta has a terminal spine, shorter 
and thicker than in pumila. 

If pumila and kayankulamensis should prove to be conspecific, the 
occurrence of the species in two such widely separated localities must 
indicate either transference from one area to the other or importation 
into both areas from still a third region, presumably somewhere in 
eastern Australasia. 

Next to kayankulamensis, pumila appears to be closest to javanica 
from which it is distinguishable by the slightly more lateral location of 
spermathecal pores, presence of hearts in xiii, more posterior origin of 
the intestine, and slightly shorter penial setae. The suggestion has 
already been advanced that javanica may be an importation into both 
Java and Burma from some eastern portion of Australasia. 

Diagnosis. Quadrithecal ; spermathecal pores on 7/8-8/9, on or 
slightly median to b. Male pores in ab, on xviii, on or close to the 
center of small, transversely placed, slightly protuberant porophores 
of shortly elliptical outline in ab or reaching just median to a or just 
lateral to b. Genital marking a transversely placed, whitened area of 
elongately elliptical outline, in aa, on 17/18. Female pore median. 
Seta (/ dorsal behind the clitellum and after the first few postclitellar 
segments close to the mid-dorsal line so that dd<aa. First dorsal 
pore on 6/7. Unpigmented. Length to 76 mm. Diameter 1 mm. 

Gizzard in v. Intestine begins in xvi or xvii. Hearts latero-oesopha- 
geal, in x-xiii. Prostatic duct J^ mm. long, straight, slender. Sperma- 
thecal duct slightly shorter than ampulla, diverticulum into lateral 
face of duct nearer ampulla. Penial setae 0.28-0.42 mm. long, 0.005- 
0.008 thick at base, 0.006-0.008 at midshaft, tip terminating in a 
hair-like filament. 



gates: peregrine earthworms 119 

Genus Pheretima Kinberg 1867 

Pheretima bicincta (E. Perrier) 

1875. Perichaeta bicincta E. Perrier, C. R. Ac. Sci. Paris, LXXXI, p. 1044. 
(Types in the Paris Museum. Type locality unknown but suppo.sed 
to be on Luzon or Mindoro, Philippine Islands.) 

Material examined. From Burmese collections: 

"Earth in large flower pots on veranda of faculty house", Judson College 
Compound, Kokine, Rangoon, 1 aclitellate and 7 clitellate specimens. 

External characteristics. The setae of x quite definitely appear to 
be smaller than those of ix and xi but have not been measured. The 
numbejrs of male setae on xviii and of setae on xx are; 8, 8, 8, 7, 7, 8, 8, 
and 47, 48, 46, 46, 48, 46, 48. The first dorsal pore is on 12/13 (8). 
Female pores are paired, diagonally placed slits, each pore very slightly 
anterior and median to a. Each male pore is at or near the center of a 
small, transversely placed, indistinctly demarcated area with outline 
approximating to shortly elliptical. 

Genital markings are unrecognizable. On several specimens a num- 
ber of minute, greyish translucent, circular areas on the postsetal 
portion of xviii median to the male pore lines are recognizable under 
best optical conditions. These areas probably mark sites of gland 
pores. , 

Internal anatomy. Septum 9/10 is strong and does not rupture 
easily but is almost transparent. On the oesophagus just anterior to 
9/10 there is a ring of slight protuberances, probably rudiments of 
glands that form a circumoesophageal collar in other species. On the 
inner wall of the gut in x-xiii there is on each side a series of vertically 
placed ridges; longitudinal ridges at the median line dorsally and ven- 
trally not noted. In xiv and the anteriormost portion of xv the gut 
is narrow and provided internally with low, longitudinally placed, 
whitish ridges. The typhlosole begins abruptly in the caecal segment 
and is a simple ridge about }/^ mm. high. Passing posteriorly the typh- 
losole gradually decreases in height and is unrecognizable posterior to 
1. There are masses of nephridia in v and vi but blood glands and 
lymph glands have not been found. Hearts of x-xii connect the ven- 
tral and supra-oesophageal vessels, branches to the dorsal trunk not 
found. 

Seminal vesicles of xii are acinous, vertically placed bodies on the 
posterior face of 11/12, the vesicles of xi also acinous but smaller, in- 
cluded in the testis sac of xi. Posterior and just median to the ectal 
end of each prostatic duct and sessile on the parietes is a soft mass of 



120 bulletin: museum of comparative zoology 

circular outline of glandular tissue which can be scraped away from 
the body wall leaving little if any evidence of its presence. There is 
no spermatozoal iridescence in the seminal chambers (4 specimens). 

Remarks. All specimens are probably in a postsexual condition 
although the clitellum has regressed on one specimen. As there appears 
to be no variation with regard to the number of female pores the bi- 
poral condition is regarded as a specific characteristic. Although genital 
markings appear to be lacking masses of glandular material in xviii 
are termed genital marking glands since indistinctly outlined genital 
markings have been noted in other specimens of the species, though 
there apparently less marked (epidermal modification slight) and 
more transitory than usual. 

P. bicincta has not been reported hitherto from Burma though 
twice recorded from India, and was not obtained in other years when 
plants were repotted. The original home of the species is still unknown. 

Pheretima humilis spec. nov. 

Material examined. From Burmese collections: 

"Earth in large flower pots on west veranda of faculty house," Judson 
College compound, Kokine, Rangoon, September, 1936, 6 clitellate 
specimens (Types). 
Same locality but after repotting, June, 1937, 10 clitellate specimens. 

External characteristics. Length 20-23 mm. Diameter two mm. 
Segments; 75 (1), 78 (1), 79 (1), 80 (3). Unpigmented. Setae begin 
on ii on which there is a complete circle (6). Setal formulae are shown 
below. The first dorsal pore is on 12/13 (6). The clitellum is brownish, 
slightly protuberant, annular, extending from 13/14 to 16/17; inter- 
segmental furrows and dorsal pores lacking, setae present ventrally 



on 


XVI. 


























Setal Formulae 










V 


vi 


xvii 


xviii 


xix 


ii Hi 


via 


xii 


xvi 


XX 


29 


33 


12 


7 


10 


40 47 


58 


52 


6 


42 


29 


29 


11 


7 


11 


40 + 


56 


48 


6 


44 


26 


27 


12 


8 


12 




55 


50 


12 


40 


.. 




10 


8 


12 


. . 


. . 


48 


11 


42 


, , * 




11 


6 


10 




54 


49 


8 


38 






12 


8 


11 


. . 


57 


46 


6 


41 



Bithecal, spermathecal pores minute and superficial, on 5/6. There 
is a single female pore on four specimens, possibly a single pore on the 



gates: peregrine earthworms 121 

other two. INIale pores have not been seen but are obviously minute 
and superficial, at or near centers of disc-like porophores of circular 
outline. No genital markings. 

Internal anatomy. None of the septa are thickly muscular; 8/9 com- 
plete but membranous, 9/10 apparently lacking. The intestine begins 
in XV (4). Intestinal caeca are simple (4). The single heart of ix is on 
the right side (1) or the left side (1), a pair of hearts belonging to ix 
present in two specimens. The last pair of hearts is in xiii (4). All 
hearts of ix-xiii pass into the ventral blood vessel (2). 

Testis sacs are probably unpaired and annular, except that the sac 
of xi in one specimen appears to be U-shaped, the limbs of the U reach- 
ing nearly to the dorsal blood vessel ; hearts of x and xi included in the 
sacs (3), the dorsal blood vessel included in the sac of x (3) or beneath 
the sac (1), beneath the testis sac of xi (2). Seminal vesicles of xi are 
small and are included within the posterior testis sac or sacs (4). Vesi- 
cles of xii are small, vertically placed on the posterior face of 11/12, 
scarcely thicker than the hearts of xii. Prostates extend through xvi or 
xvii to xix. The prostatic duct is 1/^-2 mm. long, variously twisted or 
bent into a U-shaped loop. 

The spermathecal duct is slender, longer than the ampulla (3), the 
wall thick, lumen abruptly narrowed at the diverticular junction. The 
diverticulum which passes into the median face of the duct near to 
but not at the parietes is as long as or longer than the combined lengths 
of duct and ampulla. The ectal half of the diverticulum, the stalk, 
has a very narrow lumen while the ental half is marked off into three 
rather ovoidal, equisized seminal chambers separated from each other 
by short, narrow, neck-like regions, each chamber distended by a hard, 
practically transparent mass. 

Remarks. Because of the small size, determination of characteristics 
of the testis sacs is difficult. The ventral blood vessel appears to be 
actually within the sacs but filling the lumen so that testicular coagu- 
lum is not continuous from one side to the other ventrally. If how- 
ever the ventral trunk is not within the sacs, characterization must be 
horseshoe-shaped . 

Several small bithecal species with spermathecal pores on 5/6 are 
known. From these huviilis may be distinguished as follows; from P. 
lompobatangensis (Michaelsen) 1899 by absence of genital markings 
and of 9/10, the shape of the spermathecal diverticulum; from P. 
nugalis Gates 1931 by the sliape of the testis sacs and inclusion of 
seminal vesicles; from P. ptisilla (Ude) 1893 by absence of genital 
markings; from P. voeltzkowi Michaelsen 1907 by absence of genital 



122 bulletin: museum of comparative zoology 

markings and of 9/10, the presence of hearts in xiii; from P. wui Chen 
1935 by the absence of genital markings, the shape of the testis sacs 
and inclusion of seminal vesicles; from P. zoysiae Chen 1933 by the 
presence of septum 8/9, more dorsal location of the spermathecal 
pores and the larger number of spermathecal setae. 

In spite of extensive collecting throughout Burma over a period of 
more than fifteen years and the examination of many thousands of 
specimens from all sorts of situations in and around the city of Ran- 
goon the species has not been found in a natural environment and 
disappeared from the author's flower pots after the second year of 
collection. In view of these facts and the association in the same pots 
with the obviously peregrine hicincta it is probable that humilis is also 
an importation into Burma. 

The description above is based on the first series only, in order to 
save the second series for future reference. 

Diagnosis. Bithecal, spermathecal pores minute and superficial, on 
5/6. Male pores minute and superficial, each pore on a circular, disc- 
shaped porophore. Setae present ventrally on xvi: v/26-29, vi/27-33, 
xvii/10-12, xviii/6-8, xix/10-12, 40/ii, 47/iii, 54-58/viii, 46-52/xii, 
6-12/xvi, 38-44/xx. First dorsal pore on 12/13. Length 20-23 mm. 
Diameter 2 mm. Segments 75-80. 

Septum 8/9 present but membranous. Intestinal caeca simple. 
Testis sacs annular, seminal vesicles of xi included. Spermathecal 
duct as long as or longer than ampulla, diverticulum as long as or 
longer than main axis, with long stalk and three ovoidal seminal 
chambers separated from each other by short, neck-like regions. 



Genus Ramiella Stephenson 1921 

Six species are known of which most are inadequately characterized. 
In fact there is some doubt as to the correctness of including all of 
these forms in one genus. All species are small, one to two mm. in 
diameter and therefore especially liable to accidental transportation. 



Ramiella cultrifera Stephenson 

1931 . Ramidla cultrifera Stephenson, Rec. Ind. Mus. XXXIII, p. 187. (Type 
locality Kangoon. Types in the British Museum and Judson Col- 
lege.) 

1 935. (Ramiella cultrifera, Michaelsen, Ann. Mag. Nat. Hist. (10), XV, p. 103. 
(Christmas Island near Java.) 



gates: peregrine earthworms 123 

Material examined. From Burmese collections: 

Rangoon, Hanthawaddy district, July, 1 clitellate specimen; August, 1 

aclitellate specimen. K. John. 
Hmawbj, Insein district, September, 1 aclitellate specimen. K. .John. 
Toungoo, October, 1 clitellate specimen. K. John. 
Mt. Popa, Myingyan district, September, 1 clitellate specimen. K. John. 

External characteristics. Length 33-35 mm. Diameter ca. 1 mm. 
The Rangoon worm is only 20 mm. long but may have autotomized a 
tail portion. Unpigmented. The prostomium is epilobous (tongue 
not marked off posteriorly — Mt. Popa specimen, tongue pointed 
posteriorly — Toungoo specimen, normal — Hmawbi specimen). 

The clitellum is saddle-shaped, lacking in aa (Mt. Popa specimen, 
condition not determinable on Toungoo specimen as result of dam- 
age), from 12/13 to 17/18; intersegmental furrows and dorsal pores 
lacking. 

The spermathecal pores are on viii and ix, on b, on the anteriormost 
margins of the segments just behind the intersegmental furrows. After 
dissecting the spermathecal ducts out from the parietes a strand of 
tissue is visible between the aperture and the intersegmental furrow. 
On one of the types the spermathecal pores are open, the segmental 
location readily recognizable. 

The female pores are paired, slightly anterior and just median to a. 

The prostatic and male pores are in ab, and on the setal arcs of xvii 
(or just behind), xviii and xix, perhaps a trifle nearer to 6 than a, 
the prostatic pores (common apertures of prostatic ducts and 
penisetal follicles) usually recognizable without difficulty, the 
male pores apparently a trifle smaller. The seminal grooves are nearly 
straight. 

Genital markings are tiny, slightly raised tubercles of circular to 
transversely and shortly elliptical outline, with depressed, greyish 
translucent central spot: in ab just median to the spermathecal pores 
and at the posterior margin of viii (1 type and the Mt. Popa specimen) ; 
on X, in ab, between the postsetal secondary furrow and 10/11 (Toun- 
goo specimen); lacking on four types and the aclitellate Hmawbi 
specimen. 

Internal anatomy. Septa 5/6-10/11 are muscular and relative to 
the size of the animal might be called thickly muscular; 11/12 slightly 
muscular, opaque. 

The gizzard is about twice the width and length of the portion of the 
gut in the segment next behind, "with more or less marked muscular 
sheen, in vi (Toungoo and Hmawbi specimens). The oesophagus in 



124 bulletin: museum of comparative zoology 

ix-xii is slightly moniliform, constricted septally, especially wide in 
X and xi, with low, irregular, closely crowded ridges placed vertically 
in xi but longitudinally in x, and larger, smoother, white, longitudinal 
ridges in xii-xiii. The intestine begins in xiv (4), the oesophageal 
valve in the anteriormost portion of xiv or reaching into xiii. A slightly 
irregular and low ridge which may be a rudimentary typhlosole is 
recognizable in one specimen from xvii posteriorly. 

The dorsal blood vessel (single) is continued to the region of the 
cerebral ganglia. A supra-oesophageal trunk is present in x-xiii. 
Extra-oesophageal trunks are visible in vii or viii to ix or x, but are 
empty and unrecognizable anteriorly and posteriorly. Lateroparietal 
trunks from the region of xiv-xvi pass in xiii to the gut. No subneural. 
The hearts of xii are bifid (2), one branch passing to the supra-oesopha- 
geal trunk, the other to the dorsal vessel. Hearts of xi have been 
traced to the supra-oesophageal only, of x to the dorsal trunk only. 
The last pair of hearts is in xii (3). 

Nephridia, poorly preserved and easily ruptured are flattened against 
the parietes in the postclitellar segments; in two longitudinal rows on 
each side, attached about at b and d. In one specimen there appears 
to be an additional row, mesially on each side, of much smaller ne- 
phridia. The posteriormost segments are filled with coagulum that is 
adherent to coelomic walls and segmental organs, removal of the coagu- 
lum almost impossible without damage to the excretory tubules, but 
a slender nephridial duct apparently passes into the parietes near 
seta b. Alongside this duct is a delicate filament (neck?) which can be 
traced occasionally to and through the septum next in front where it 
is enlarged into a funnel-like structure close to the ventral parietes. 
The neck and funnel break off and are lost with such ease that all at- 
tempts to remove and mount them for microscopic examination have 
been unsuccessful. 

Male funnels are present in x and xi, both pairs with brilliant 
spermatozoal iridescence (Toungoo specimen), or the anterior funnels 
with no iridescence and the posterior funnels slightly iridescent (Mt. 
Popa specimen). The seminal vesicles are paired in xi and xii (Mt. 
Popa specimen) or present in xii only (Hmawbi and Toungoo speci- 
mens), in the Popa specimen reaching into contact with the dorsal 
blood vessel. The vas deferens of the Popa specimen is visible as a 
slightly irregular but strongly iridescent filament on or in the parietes 
and can be traced back to xviii where it disappears into the parietes 
midway between the prostatic ducts of a side, in xvii passing lateral 
to the prostatic ducts. The prostates are slightly flattened, shortly 



gates: peregrine earthworms 125 

elliptical in transverse section, looped ; the anterior pair reaching into 
xviii, the posterior pair reaching into xx. The prostatic ducts are 
short, usually with slight (muscular?) sheen. 

Two penisetal follicles, each containing two setae, pass into the 
parietes on the median face of each prostatic duct in the Popa speci- 
men. Presumably one seta in each follicle is to be regarded as a prac- 
tically mature, reserve seta. In the Hmawbi and Toungoo specimens 
there appears to be but a single follicle, each containing two setae, 
associated with each prostatic duct but the tissues are very delicate and 
readily separate leaving strands attached to each seta that may repre- 
sent portions of two distinct follicles. In the Popa specimen both 
setae of any follicle maybe alike or of the two types figured by Stephen- 
son. In the Toungoo specimen the two types of setae are associated 
with each prostatic duct. The supposed differences between the two 
types of setae appear to be fictitious and of no importance. Actually 
each seta is ribbon-like, with the lateral margins along the whole of the 
shaft, or only a part, rolled into contact. Stephenson's figures 10 and 
13 show a complete rolling, while figures 9 and 12 show a rolling of an 
ectal portion of the shaft. In figure 9 one margin only is rolled while 
in figure 12b both margins are slightly rolled, the bottom margin more 
so than the other. Pressure on the setae will usually flatten out most 
of the rolled portion of a shaft. Three setae mounted in water were 
flattened out or unrolled merely by the cover glass pressure induced 
by the evaporation of water while an earlier mount was being ex- 
amined. The longitudinal line along the shaft towards the right in 
figure 13 represents the region where the edges or margins come into 
contact or overlap. This line is usually not to be seen, presumably 
because it runs along the side of the shaft as the seta lies on the slide 
but on one seta this line runs along the upper side of the shaft and is 
visible as a very narrow groove from the base of the tip almost to the 
ental end. When the ectal ends have been partially unrolled a short 
region ental to the point where the margins first meet or overlap is 
quite clearly cylindrical but further entally no trace of a hollow or cen- 
tral cavity is visible and the shaft appears to be solid. Each seta is 
bent in an arc as shown by Stephenson. A terminal ectal portion about 
0.03 mm. long is abruptly narrowed, wrinkled and hooked to one side. 
This part appears to be solid and cannot be flattened out or unrolled. 
The ornamentation is of 7-15 short, transverse rows of triangular 
teeth of variable size, the rows not reaching across the whole of the 
shaft, and unrecognizable as rows until the shaft is unrolled or flat- 
tened. In addition, just behind the tip there may be a few scattered 



126 



bulletin: museum of comparative zoology 



teeth or irregular rugosities. Measurements in mm. (by Miss Chapman) 
are given below. 

Penial setae 









Width at base 


Width at midshaft 










after 


before 


after 


before 




Segment 


Seta 


Length 


unrolling 


unrolling 


unroll 


ing 


unrolling 


Locality 


xvii 


a 


0.66 


0.030 










Mt. Popa 


xvii 


a 


0.59 


0.024 












xvii 


h 


0.50 


0.028 


0.012 










xvii 


h 


0.58 


0.028 


0.014 










xvii 


a 


0.60 


0.020 


0.016 




. 


.... 




xvii 


a 


0.61 


0.027 


0.015 






.... 




xvii 


b 


0.58 


0.028 








.... 




xvii 


b 


0.52 


0.025 








.... 




xix 


a 


0.74 


0.022 


0.014 






* . > . 




xix 


a 


0.56 


0.022 


0.015 






.... 




xix 


a 


0.57 


0.024 


0.019 










xix 


b 


0.58 


0.028 


t 






.... 




xix 


b 


0.60 


0.022 


.... 










xvii 


a 


0.93 


0.033 


0.024 


0.022 


0.011 


Toungoo 


xvii 


b 


0.84 


0.034 




0.027 


.... 




xix 


a 


0.95 


0.032 


0.024 


0.024 


0.013 




xix 


b 


0.82 


0.036 




0.028 






• . . 




0.95 


0.031 


0.022 


0.030 


0.017 


Hmawbi 


. . . 




0.85 


0.034 


0.024 


0.031 


0.016 




. . . 




0.92 


0.032 


0.026 


0.029 


0.016 








0.83 




0.027 


0.03 





0.017 





The ovaries contain a few relatively large ova. In the Toungoo 
specimen there is a pair of small ovisacs in xiv. 

The spermathecal duct is slender, longer than the ampulla, erect 
and straight, nearly circular in cross section, the lumen narrow and 
slightly irregular, the wall thick, the epithelium lining the lumen high. 
The unstalked diverticulum may be borne dorsally at the ental end of 
the duct with the ampulla hanging ventrally, or the diverticulum and 
the ampulla may be perpendicular to the end of the duct and on op- 
posite sides, or both diverticulum and ampulla may be pendent on 
opposite sides from the ental end of the duct. In the aclitellate 
Hmawbi specimen diverticulum and ampulla are of about the same 
size, in other words the ampulla larger. Each diverticulum of the 
clitellate specimens is characterized by a spermatozoal iridescence, the 
spermatozoa in a long cord which is twisted and looped in a com- 



gates: peregrine earthworms 127 

plicated but quite irregular fashion within the diverticulum. In two 
diverticula the spermatozoal cord passes out of the diverticulum and 
straight across the duct lumen and down into the ectal portion of the 
ampulla. The latter is filled with a whitish, non iridescent material. 

Remarks. Spermathecal conformations like those shown by Stephen- 
son in fig. 8 or by Michaelsen in fig. 1 have not been found, the ampulla 
of the Christmas Island worm usually elongated. Nor has any e^adence 
of Michaelsen's seminal chambers been found in the diverticula. The 
margins of the spermatozoal cord are clearly visible, the coils of the 
cord apparently filling the whole of a single-chambered diverticulum. 

Stephenson found a gizzard in vii in one Rangoon specimen, and a 
similar location was noted for the Popa specimen but could not be 
verified after study of later specimens as the worm had been dis- 
carded. Michaelsen found the gizzard in vi (the usual location in the 
genus) in his specimens from Easter Island. 

In addition to an apparent intraspecific variation in the segmental 
location of the gizzard there is variation in the conformation of the 
prostomium, completeness of the clitellum ventrall}^ location of the 
first dorsal pore, setal ratios, presence or absence of ovisacs and the 
angle of the spermathecal ampulla to the duct. At present these 
variations do not appear to be of especial importance (the incomplete- 
ness of the clitellum mid-ventrally may be due to beginning regression) 
but most species of the genus are very much alike, distinguished from 
each other at present rather unsatisfactorily by characteristics of 
(inadequately described?) penial setae and spermathecae and by the 
numbers of nephridia. 

Diagnosis. Spermathecal pores on, b, just behind 7/8 and 8/9. 
Prostatic pores (conjoined openings of prostatic ducts and penisetal 
follicles) in ab at termini of seminal grooves that reach from setal arc 
of xix to or nearly to the arc of xvii. Male pores in the grooves midway 
between the prostatic pores. Genital markings (when present) post- 
setal in ab, on viii and x. Clitellum annular, on xiii-xvii. First dorsal 
pore on 6/7-8/9. Length 20-35 mm. Diameter 1-1.2 mm. 

Gizzard in vi. Holandric; seminal vesicles in xii. Spermathecal duct 
slender and longer than the ampulla; diverticulum spheroidal to 
ellipsoidal, sessile on ental end of duct. Penial setae ribbon-like, but 
with lateral edges rolled together along a considerable portion or the 
whole of the shaft, with a narrowed, hooked, and solid (?) tip; 0.50-0.95 
mm. long, 0.012-0.027 thick at base unflattened, 0.022-0.034 flattened, 
0.011-0.017 at midshaft unflattened or 0.022-0.031 flattened. 



128 bulletin: museum of comparative zoology 

Genus Dichogaster Beddard 1888 

Although supposedly preserved in the usual manner much of the 
Burmese Dichogastrid material has been found to be in unsatisfactory 
condition perhaps because of maceration due to overcrowding in 
storage tubes. When worms of such small size are softened internally, 
determination of external characteristics, at best a rather tedious task, 
may be more than usually difficult or even impossible. Coelomic 
cavities are often filled with a sticky coagulum adherent to nephridia, 
ovaries, etc., and in washing out the coagulum considerable damage 
may be done to internal structures. 

Accordingly a complete account of the internal anatomy of each 
species is impossible and results of studies of two forms in particular 
are so unsatisfactory that they have been discarded almost in toto. 

In species of such small size external characteristics that will enable 
identification are especially needed. As a result of the study of the 
material listed below it may be stated that Burmese species at least 
can be recognized by the female pores (number and location) alone or 
in connection with certain easily recognized external characteristics. 
It is perhaps unnecessary to point out that for the present at least, 
any identification made from external characteristics should be con- 
firmed from examination of internal structures with particular atten- 
tion to the penial setae. 



Dichogaster affinis (Michaelsen) 

1890. Benhamia affinis Michaelsen, Mitt. Mus. Hamburg, VII, p. 9. (Type 
locality Quilimane, Zanzibar. Type in the Hamburg Museum.) 

Material examined. From Burmese collections: 

Bilin, Thaton district, September, 1 clitellate specimen. K. John. 
Duyinzeik, Thaton district, September, 1 clitellate specimen. K. John. 
Kyaikto, Thaton district, September, 7 clitellate specimens. K. John. 
"Jungle", west of Pegu, Pegu district, August, 2 clitellate specimens. K 

John. 
Thanbula, Thayetmyo district, August, 1 clitellate specimen. K. John. 
Magwe, Magwe district, August, 2 clitellate specimens. K. John. 
Taungdwingyi, Magwe district, August, 6 clitellate specimens. K. John. 
Pyinmana, Yamethin district, October, 5 clitellate specimens. K. John 
Mt. Popa, Myingyan district, September, 3 clitellate specimens. K. John. 
Dwehla, Kyaukse district, September, 2 clitellate specimens. K. John. 
Myotha, Sagaing district, September, 21 cUtellate specimens. K. John. 



gates: peregrine earthworms 129 

Taungyi and Lashio, F. S. S., August-September, 1 aclitellate and 30 

clitellate specimens. H. Young. 
Kyaukmyaung, Shwebo district, August, 1 clitellate specimen. Saw San 

Thwe. 
Tiangzup, Myitkyina district, November, 13 clitellate specimens. F. D. 

Forbes. 

External characteristics. Female pores are immediately in front of a. 

Genital markings, unpaired and median, are located as follows: 
on 8/9, 72 specimens (1 from Kyaukmyaung, 1 from Thanbula, 1 from 
Bilin, 2 from Dwehla, 1 from Duyinzeik, 4 from Kyaikto, 3 from 
Pyinmana, 6 from Taungdwingyi, 2 from Pegu, 18 from Myotha, 12 
from Tiangzup, 21 from Lashio and Taungyi); on 8/9 and 9/10, 15 
specimens (1 from Mt. Popa, 2 from Magwe, 1 from Pyinmana, 3 from 
Myotha, 7 from Lashio and Taungyi, 1 from Tiangzup); on 9/10, 6 
specimens (1 from Mt. Popa, 3 from Kyaikto, 2 from Lashio); on 
9/10-10/11, 2 specimens (from Pyinmana and Mt. Popa) ; on 7/8-9/10, 
1 specimen (Lashio). 

DiCHOGASTER BOLAUi (Michaelsen) 

1891. Benhamia holavi Michaelsen, Mitt. Mus. Hamburg, VIII, p. 307. (Type 

locality Eergedorf, Hamburg. Types in the Hamburg Museum.) 
Material examined. From Burmese collections: 

Sittang, Thaton district, October, 2 clitellate specimens. K. John. 

Kyaikto, Thaton district, September, 5 clitellate specimens. K. John. 

Frome, Frome district, August, 1 chtellate specimen. K. John. 

Toungoo, Toungoo district, October, 10 clitellate specimens. K. John. 

Thanbula, Thayetmyo district, August, 1 chtellate specimen. K. John. 

Minbu, Minbu district, August, 6 chtellate specimens. K. John. 

Taungdwingyi, Magwe district, August, 1 chtellate specimen. K. John. 

Fyinmana, Yamethin district, October, 17 clitellate specimens. K. John. 

Mt. Fopa, Myingyan district, September, 4 clitellate specimens. K. John. 

Taungtha, Mjangyan district, September, 3 chtellate specimens. K. John. 

Dwehla, Kyaukse district, September, 1 chtellate specimen. K. John. 

Myotha, Sagaing district, September, 4 chtellate specimens. K. John. 

Civil Lines, Mandalay, Mandalay district, October, 3 achtellate speci- 
mens. Miss M. Chapman. 

Kyaukmyaimg, Shwebo district, August, 1 clitellate specimen. Saw San 
Thwe. 

Katha, Katha district, August, 1 clitellate specimen. Saw San Thwe. 

Taungyi and Lashio, F. S. S., August, several chtellate specimens. H. 
Young. 



130 bulletin: museum of comparative zoology 

Wasat Hka, Myitkyina district, November, 1 clitellate specimen. F. D. 

Forbes. 
Tingpai, Mjatkyina district, November, 7 clitellate specimens. F. D. 

Forbes. 

External characteristics. The spermathecal pores are on or close to a. 
The single female pore is median. 

DiCHOGASTER MODIGLIANII (Rosa) 1896 

1896. Benhamia modiglianii Rosa, Ann. Mus. Genova, XXXVI, p. 510. 
(Type locality Padang, Sumatra. Type in the Genoa Museum?) 

1931. Dichogaster modiglianii, Stephenson, Proc. Zool. Soc. London, 1931, 
p. 65; Rec. Ind. Mus. Calcutta, XXXIII, p. 198; — D. doveri Stephen- 
son, J. Fed. Malay States Mus., XVI, p. 276. (Type locahty of 
doveri Kuala LunTpur, Selangor, F. M. S. Types in the British Mu- 
seum.) 

Material examined. From Burmese collections: 

Sittang, Thaton district, October, 3 clitellate specimens. K. John. 
Kyaikto, Thaton district, September, 1 clitellate specimen. K. John. 
Kayan, Hanthawaddy district, September, 2 clitellate specimens. K. John. 
Twante, Hanthawaddy district, September, 6 clitellate specimens. K. 

John. 
Rangoon, Hanthawaddy district, January, 4 clitellate specimens. K. 

John. 
Magwe, Magwe district, August, 1 clitellate specimen. K. John. 
Toungoo, Toungoo district, October, 1 clitellate specimen. K. John. 
Taungdwingyi, Magwe district, August, 3 clitellate specimens. K. John. 
Pyinmana, Yamethin district, October, 12 clitellate specimens. K. John. 
Mt. Popa, Myingyan district, September, 7 chteUate specimens. K. John. 
Hills near Naba, Katha district, September, 8 cUtellate specimens. Saw 

San Thwe. 

External characteristics. The prostomium is proepilobous, segment 
i divided mid-dorsally by a longitudinal groove to 1/2 which is normally 
developed. Setae begin on ii. The first dorsal pore is probably on 5/6 
but on 4/.5 of some of the specimens there is a pore-like marking which 
may be perforate. The clitellum is annular but the epidermal thicken- 
ing is less developed in aa though this is only recognizable in trans- 
verse incisions through the body wall. 

. Spermathecal pores are on or very close to a. On each specimen the 
female pores are on the setal arc or (occasionally) just a trifle behind, 
and just median to a. 



gates: peregrine earthworms 131 

Seminal grooves are nearly straight; on or very close to a, penial 
setae protuberant to the exterior from the anterior and posterior ends 
of the grooves where the prostatic apertures presumably are located. 
The male pores are represented by dark spots about midway between 
sites of the prostatic apertures. A longitudinally placed, median and 
almost rectangular area including the seminal grooves is often slightly 
whitened but not protuberant. 

Internal anatomy. Gizzards are large, always anterior to 8/9 and 
presumably in vii and viii (13), no septum recognizable between the 
gizzards. Calciferous glands are in xv-xvii (13). The intestine begins 
in xix (13), the oesophageal valve in xviii or just reaching into xix (2). 
The typhlosole which begins more or less abruptly in xxii-xxiii (10) is 
a high, simple lamella, bent back and forth rather regularly, an anterior 
portion opaque, the posterior half rather translucent. In worms with 
110-120 segments the typhlosole ends rather abruptly in Ixxviii-lxxxi 
(10). In one specimen with only 84 segments the typhlosole ends 
abruptly in Ixviii. In another worm, even shorter, the typhlosole is 
continued to the hind end of the gut. Just lateral to the median typhlo- 
sole, on each side, and in four or five successive segments beginning at 
xxiii or xxiv there is a lateral typhlosole, a low but quite definite, 
scarcely lamelliform ridge (13) which is interrupted or very low and 
scarcely recognizable midsegmentally in each segment. 

The dorsal blood vessel (single) is continued anteriorly to the region 
of the supra-pharyngeal ganglia. The ventral trunk is first recogniza- 
ble at the anterior margin of the subpharyngeal ganglia. A supra- 
oesophageal trunk is usually not recognizable and when visible can be 
seen only in x-xiii. Extra-oesophageal trunks are first visible in the 
region of the circumpharyngeal nervous commissures from where they 
run posteriorly in the mass of tissue on and behind the pharynx, recei v- • 
ing several large dorsal branches, with a large transverse connective 
just anterior to the first gizzard, the trunks well below and lateral to 
the gizzards, passing onto the ventral face of the gut in ix from whence 
they gradually approximate to the midventral line until in xii they are 
actually in contact, unrecognizable posterior to 12/13. In several 
specimens a parietal vessel in xiv-xix on each side passes up onto the 
anterior face of 13/14 and apparently into the supra-oesophageal. 
Anteriorly in xii the extra- and supra-oesophageals are connected by 
fairly large commissures running around the gut; a smaller pair anter- 
iorly in xi, a still smaller pair on the posterior face of 9/10. The last 
pair of hearts is in xii (13), the hearts of x-xii bifurcating dorsally, one 
branch apparently passing into the supra-oesophageal and the other 



132 bulletin: museum of comparative zoology 

to the dorsal trunk though it has not been possible to trace both 
branches of any one heart to connections with the main trunks. All 
hearts pass into the ventral vessel. 

Nephridia, from xx or xxi posteriorly, are fairly large discs flattened 
against the parietes, at times almost transversely rectangular, in con- 
tact or almost so with both septa of a segment, slightly yellowish and 
with a translucent granular appearance, tubules unrecognizable or 
recognizable only at the anterior margins of the discs in xx or xxi. 
These nephridia are in four longitudinal rows on each side, but the 
medianmost nephridium is quite small, in contact with or actually 
attached to the nephridium next laterally of which it appears to be a 
small appendage. From xx or xix to xiii the nephridia have little or no 
granular investment and are opaque, whitish loops, located in the pos- 
terior portions of the segments, just in front of the septa. From xii 
anteriorly nephridia, in pinned out specimens, are on the anterior faces 
of the septa. 

Septa 10/11 and 11/12 are in contact peripherally to form a testicular 
chamber that is not opened in carefully dissected specimens. In several 
worms this chamber, perhaps unusually distended by the testicular 
coagulum, is herniated anteriorly for some distance along the nerve 
cord. Beneath the gut in xi is a closed off, transversely placed and me- 
dian testis sac. In one specimen a herniation from the posterior sac 
passes forwards into the anterior herniation of the sac of x. Sacs of x 
and xi are usually filled with coagulum, all male funnels characterized 
by a brilliant spermatozoal iridescence. iVlthough the worms are 
sexually functional, seminal vesicles have not been found though tiny 
whitish structures on the posterior face of 11/12 may represent rudi- 
ments of vesicles. Prostates are short and almost straight, slightly 
flattened, usually confined to xvii and xix, often attached to the anter- 
ior faces of 17/18 and 19/20, occasionally penetrating slightly into 
xviii and xx. The ducts are slenderer than the glands, with slight mus- 
cular sheen. Male deferent ducts are easily traced and pass into the 
parietes in xviii midway between the ectal ends of the prostatic ducts 
of a side. A column of muscular and connective tissue on the median 
face of each prostatic duct contains the penial setae. With care it is 
possible to separate out from this column two follicles (?) each contain- 
ing one penial seta. In one column, after the supposed follicles contain- 
ing the functional setae had been removed, two setae were still visible. 
These are nearly grown reserve setae, one of each type. Close to the 
base of each reserve seta is a large, ellipsoidal, vesicular nucleus 
within which a single, spheroidal endosome is visible. The penial setae 



gates: peregrine earthworms 133 

are like those of dnirri Steplienson 1981 but there are usually five to 
seven teeth visible at each margin of the larger type. 

The spermathecal ampulla is shorter than the duct, occasionally 
much shorter, shortly ellipsoidal and clearly marked off. The duct is 
slightly bulbous, widened in a middle portion and not especially nar- 
rowed in the muscular layers of the body wall, the lumen narrow ectally 
and with smooth wall, widened in the middle portion, then narrowed 
again entally though not quite as narrow or with as smooth wall as 
ectally. The diverticulum comprises a small, spheroidal to shortly 
ellipsoidal seminal chamber which may be erect or pendent, and a 
very short and slender stalk that passes into the anterior face of the 
duct at or just lielow the region of greatest thickness, opening upwards 
into the middle chamber of the duct. The seminal chamber is usually 
characterized by a brilliant iridescence. In a number of spermathecae 
the middle chamber of the duct is practically filled by a shortly ellip- 
soidal mass of spermatozoa which is much larger than that of the sem- 
inal chamber. In se\eral ampullae there are clumps of rod-like struc- 
tures that may be crystals or possibly fragments of penial setae. These 
rods may be transversely segmented. 

In xiv there is a pair of small whitish vesicles (13) probably ovisacs. 

Remarks. The dorsal and ventral trunks, the extra-oesophageals 
from 9/10 anteriorly and the hearts are always filled with blood anrl 
hence can be easily traced but supra-oesophageals, posterior portions 
of the extra-oesophageals and other vessels are usually quite unrecog- 
nizable in whole or in part. The supra-oesophageal and extra-oesopha- 
geal trunk and their connectives are distended with blood so that they 
can be traced in only two worms. Behind the gizzards 8/9-14/15 are 
closely crowded but can be identified by the organs between. A satis- 
factory determination of the gizzard segments and of the anterior septa 
has not however been made. 

The lateral typhlosoles are like the lateral ridges in holandric and 
primitive metandric species of Eutyphoeus but here more extended. 

In addition to the material listed above several Burmese specimens 
identified by Stephenson have been studied. Some of these are labelled 
modiglianii, others doveri. All are similar to the worms described above. 
D. doveri was erected for material from the Malay Peninsula but Ste- 
phenson later came to the conclusion that doveri must be suppressed. 
There are several mistakes in the account of doveri. The last hearts 
are in xii, not xi, the calciferous glands are in xv-xvii not xiv-xvi, and 
the intestine begins in xix not xvii. An empty testicular chamber is 
easily mistaken for a slightly thickened septum and segmental recog- 



134 bulletin: museum of comparative zoology 

nition is very difficult if ovaries and nephridia are washed out in freeing 
the specimen from coelomic coagulum. Septum 18/19 is often pushed 
anteriorly into contact with 17/18 so that the intestine appears to be- 
gin in xviii, the true relationship of the parts ascertainable only with 
very considera])le care in dissection. 

DiCHOGASTER SALIENS (Beddard) 

]<S93. Microdrilus salicns Beddard, Proc. Zool. Soc. London, 1892, p. 683. 
(Types in the British Museum ? Type locahty undesignated, original 
specimens from Singapore, Penang and Java.) 

Material examined. From Burmese collections : 

Toungoo, Toungoo district, October, 1 aclitellate and 4 olitellate sjjecimens. 
K. John. 

Lashio and Taungyi, F. S. S., September, 11 olitellate specimens. H. 
Young. 

Wasat Hka, Myitkyina district, November, 8 clitellate specimens. F. D. 
Forbes. 

Tingpai, Myitkyina district, November, 54 clitellate specimens. F. D. 
Forbes. 

Sumprabum, Myitkyina district, November, 3 clitellate specimens. F. D. 
Forbes. 

"Manure", Mythonkha, Myitkyina district, November, 3 clitellate speci- 
mens. F. D. Forbes. 

From Malayan collections: 

Cameron Highlands, Pahang, F. M. S., 1 aclitellate and 15 clitellate speci- 
mens. Raffles Mu.seum. 

External charactcrisiics. Length to 68 mm. The diameter of the larg- 
est worms in the clitellar region may be as much as 234 mm. though not 
more than two mm. elsewhere. Unpigmented. The prostomium is 
proepilobous but there is a definite furrow at the mid-dorsal line from 
the apex of the prostomium to or nearly to 1/2 or the site of 1/2. Inter- 
segmental furrow 1/2 is usually almost wholly lacking, unrecognizable 
until after removal of the cuticle and then only faintly indicated for a 
short distance on the dorsum near the median line. The first segment 
accordingly appears to be setigerous but is quite definitely longer than 
the second setigerous segment (iii) unless the prostomium is deeply 
retracted. 

All setae of ii are present; on xxi ah ca. = c(l, aa usually ca. = hc, but 
some variation. Ventral setae of xviii are lacking even on aclitellate 
specimens except as follows: a and h of left side present (3, from Ting- 
pai), a and }> of right side present (2 from Lashio), a of right side 



gates: peregrine earthworms 135 

present (1, Tingpai), b of left side present (1, Tingpai). When present 
setae are of the ordinary sigmoid type. 

The first dorsal pore is on 4/5 (1), on 5/6 (19, but with a pore-like 
though apparently imperforate marking on 4/5 of three specimens), 
on 6/7 (1). 

The clitellum is red, usually protuberant, extending from 12/13 
onto xix, to 19/20 or onto xx, annular except on xiii, xviii-xx but 
thinner in aa; intersegmental furrows lacking, dorsal pores lacking ex- 
cept on 19/20, setae present. 

Spermathecal pores are on 7/8-8/9, on or close to a, each pore on a 
tiny tubercle. Female pores are just median to a, on or just behind the 
setal arc, nearer to a than is b. 

The male genital shield is a transversely placed, almost diamond- 
shaped area, reaching laterally into be, and possibly slightly onto xvi 
and xviii (16/17 and 17/18 lacking ventrally), sharply demarcated ex- 
cept in aa. On each shield there are two, rather conical protuberances 
the highest portions of which are about in the region of ab, a slight 
longitudinal groove usually recognizable at the mid-ventral line. 
Seminal grooves are nearly straight, on the posterior faces of the pro- 
tuberances, about on a, the anterior ends about at the setal arc, the 
posterior ends approximately at site of 17/18. Male pores, definitely 
recognized on three specimens only, are at the hind ends of the seminal 
grooves and hence about on a, at or close to site of 17/18. Prostatic 
pores (usually unrecognizable and always so if penial setae are pro- 
tuberant to the exterior) are at the anterior ends of the seminal grooves 
and hence at a, on the setal arc. A single penial seta may be pro- 
tuberant from the anterior end of a seminal groove. 

The single genital marking, when present, is on 15/16, in aa, with 
a wide, opaque marginal band and a greyish translucent central area, 
outline circular or shortly elliptical and then transversely placed. 
Markings are present on 18 worms; six from Lashio and Taungyi, six 
from Wasat Hka and six from Tingpai. Of sixteen Malayan worms, 
six have the marking. 

Internal anatomy. Gizzards are anterior to 8/9 and presumably in 
vii and viii, 7/8 lacking or unrecognizable in dissections. Calciferous 
glands are one pair, each gland with three more or less reniform lobes 
in xv-xvii, the size of the lobes increasing from xv posteriorly. The 
common stalk of the three lobes of a side opens by a very small, slit- 
like aperture into the gut just lateral to the supra-oesophageal in xv. 
The intestine begins in xix just behind 18/19 (4). The typhlosole 
which begins abruptly in xxii (1) or xxiii (3) is a simple lamella, ending 



136 bulletin: museum of comparative zoology 

abruptly in Ixxxvi (1) or Ixxxviii (2, one specimen with 119 segments). 
Slightly lateral to the median typhlosole, in xxiii-xxix on each side, 
there is a lateral typhlosole, high intersegmentally, low or interrupted 
midsegmentally. 

The dorsal blood vessel (single) is continued anteriorly onto the 
pharyngeal bulb. A supra-oesophageal is usually recognizable in xi-xii 
only. Extra-oesophageals are unrecognizable behind xi, connected by 
transverse commissures just behind 6/7, anterior to 6/7 continued 
forwards close to the ventral parietes into ii. No subneural. A latero- 
parietal vessel is recognizable on each side in xiv-xvii, bifurcating on 
the anterior face of 13/14, one branch passing to the extra-oesophageal, 
the other to the supra-oesophageal. The last hearts are in xii (6). 

Nephridia in the postclitellar portion of the body are in four 
longitudinal rows on each side, each nephridium a flattened disc on the 
parietes in contact with both septa of a segment and with transparent 
granular investment. The medianmost nephridium on each side de- 
creases in size from just behind the clitellum posteriorly. In the last 
twenty segments the median nephridium of each side has a small, 
preseptal funnel. 

Seminal vesicles are represented only by small rudiments on the 
posterior face of 11/12. Prostates are restricted to xvii, the duct with 
muscular sheen, slightly looped once or twice, three quarters to one 
mm. long. The vas deferens is widened posteriorly, iridescent, readily 
visible in xv-xvii where it is looped, in xvii lateral to the prostatic duct 
and passing into the parietes behind the prostatic ducts. 

The spermathecal duct is longer than the ampulla, occasionally only 
slightly so. The ectal half of the duct has a thick wall and narrow 
lumen lined with cuticle (?). In the ental half the lumen is widened 
and more or less irregular. The diverticulum comprises a spheroidal 
to shortly ellipsoidal seminal chamber and a short, slender stalk passing 
to the anterior face of the duct. 

Penial setae are in two follicles, those of the b setae smaller, the a 
setae with more marked sinuosities and slightly thicker. Reserve 
setae are present in each follicle examined. 

The genital marking gland has a definite but thin, capsular wall," the 
longitudinal musculature thin, transparent over the gland and bulged 
upwards slightly above the general level of the parietes. A thin greyish 
spot in the body wall is however all that is visible internally without 
some dissection among muscle strands. 

Remarks. The anteriormost portion of the body is usually softened 
so that recognition of the first intersegmental furrow, even if present, 



gates: peregrine earthworms 137 

would be difficult. The furrow is however clearly undeveloped, except 
as noted, on several specimens that are almost perfectly preserved. 

Family EUDRILIDAE 

Genus Eudrilus E. Perrier 

EuDRiLUS EUGENIAE (Kinberg) 

1867. Lumbricus eugeniae Kinberg, Ofv. Ak. Forh. XXIII, p. 98. (Type 

locality St. Helena. Type in the Stockholm Museum.) 
Material examined. From Ceylonese collections: 

Heneratgoda, 40 feet, 56 juvenile or aclitellate and 14 clitellate specimens. 
The Colombo Museum. 

From American collections: 
Soil, St. Croix, Virgin Islands, 3 aclitellate and 7 clitellate specimens. 
H. A. Beatty per U. S. Nat. Mus. 

External characteristics. Length 90-130 mm. Diameter 6 mm. The 
dorsum is dark red anteriorly, gradually fading out in the posterior 
half of the body, but with a pronounced blueish tinge anterior to the 
clitellum; pigmentation, associated with the circular muscular layer, 
red. Ventrum unpigmented. The prostomium is epilobous, ca. 3^, but 
there is no transverse furrow at the posterior end of the tongue (15). 
No dorsal pores. 

Setae begin on ii on which all four couples are present; on xxiii, 
ab = cd, be ca. = ]A^aa, dd> 3^C but only sUghtly. All setae anteriorly 
and at the posterior end are sigmoid and ornamented near the tips 
with short, transversely placed, jagged ridges. 

Nephropores begin on iv (25), and are on the anterior margins of the 
segments, on or just lateral to c. Nephropores are unrecognizable on 
xviii (15 clitellate specimens) or visible but noticeably smaller than 
on x\di and xix (4 clitellate and several aclitellate specimens). 

The chtellum extends from 13/14 to 18/19 and ventrally into a 
lateral portion of aa on each side and is reddish; intersegmental fur- 
rows slightly indicated, setae usually present, except the ventral 
couples of xvii which are always lacking (even on smallest juveniles). 
The lateral setae of xiv may be lacking (1), or the ventral setae (1), 
while all setae of xiv-xvii except a and h on the left side of xvi and the 
right side of xiv are lacking on one worm. The ventral margin of the 
clitellum is only marked off externally by a disappearance of the red 
clitellar color, the epidermis gradually thinned passing ventrally. 

The spermathecal apertures are transversely placed, presetal slit 



138 bulletin: museum of comparative zoology 

on xiv, of about the same width as cd, with centers usually on or just 
median to c, occasionally lateral to c, the openings just in front of the 
setal arc. Margins of apertures are smooth or fairly so even when 
slightly protuberant. 

Apertures of the copulatory chambers are transversely placed, on 
xvii, just in front of 17/18, with centers on or lateral to h, reaching 
mesially to a, the margins slightly protuberant, whitish and finely 
lobed. The male pore is doubtless minute and was not found, probably 
on or near the tip of the penis. The latter is four mm. or more in length 
and gradually thickened dorsally, when completely retracted curved 
in a crescentic fashion, when protruded nearly straight. The aperture 
of the Y-gland is a tiny, almost minute slit on a porophore which may 
be protuberant to the exterior along with the penis. The porophore is 
much shorter than the penis, with a bluntly rounded ventral end, 
widened passing dorsally and narrowed again close to the parietes. On 
the lateral face of the porophore slightly below the ventral end there is 
a tiny, almost conical protuberance bearing at its tip the vertically 
placed, slit-like aperture of the Y-gland. From a point just above the 
pore of the Y-gland a groove runs upwards on the lateral face of the 
porophore, dorsally turning laterally to pass onto the base of the penis, 
along the median face of which it is continued to or almost to the 
ventral end. 

Definite genital markings are lacking but on each of segments xvii 
and xviii of several of the clitellate specimens, in a median portion of 
aa, there is a transversely placed, very slightly protuberant and 
rather distinctly demarcated area of shortly elliptical outline, reaching 
anteroposteriorly nearly to the intersegmental furrows. The epidermis 
of these areas is slightly thickened. 

In cd or a region reaching slightly median to c and on each side there 
is usually present a longitudinal row of tiny, circular spots. Each of 
thesje spots is of about the same size and appearance (under low powers 
of the binocular) as an open nephropore. With brilliant illumination 
and high magnification no pore is recognizable but each area is slightly 
depressed (epidermis thin) and with a brownish appearance. The 
markings are always postsetal, usually fairly close to the setal arcs, 
two per segment generally, though an extra marking or even two may 
be present on any segment and when present not far from the normal 
markings. Locations of the markings on several specimens selected at 
random are as follows: i-x (3), i-xi (1), i-xii (3), ii-x (1), ii-xi (1), 
ii-xii (1), iv-x (1), iv-v (1), extra markings not recorded. 

Iniernal anatomy. All septa from 4/5 posteriorly are present; 



gates: peregrine earthworms 139 

4/5-5/6 transparent, 6/7 and several following septa with muscular 
fibres but thin and translucent. 

The gizzard is in v (13). The oesophagus in iv is wide, narrower in 
vi-xiv, especially so in x-xii. The inner wall of the gut in vi-xiv is 
provided with low, longitudinally placed, rather irregular, whitish 
ridges. Unpaired ventral calciferous glands are large, spheroidal to 
•ovoidal, with dark surfaces, stalks very short, slender and straight, to 
ventral face of gut in median vertical plane. Paired calciferous glands 
are smaller, always white, anteroposteriorly flattened, in the posterior 
portion of xii, adherent to the lateral faces of the gut as well as to the 
anterior face of 12/13. From about the middle of the posterior face of 
each gland a stalk is continued to the anterior face of 12/13 and into 
the gut at or just behind the region of septal attachment. Small circu- 
lar openings into the median glands are readily recognizable but the 
still smaller openings of the paired glands are harder to find or recog- 
nize. Lamellae, in both paired and unpaired pouches, are vertical, in 
contact centrally in each gland so that the lumen — except just at the 
stalk region- — is reduced to small slit-like spaces. The intestine begins 
in xiv posteriorly or in xv immediately behind 14/15. In three speci- 
mens the valve is in the region of attachment to the gut of 14/15 and 
is tightly closed, the oesophageal portion of the gut in xiv clearly 
differentiated from the intestinal portion in xv. In other specimens 
the valve is relaxed and the portion of the gut in xiv is widened so that 
determination of the segment of intestinal origin is difficult or impos- 
sible. In four specimens septum 14/15 is attached to the gut behind 
the anterior end of the intestine and cannot be peeled off from this 
anterior portion. In these worms the intestine appears (o begin quite 
definitely in a posterior portion of xiv. No typhlosole. In a region ex- 
tending from approximately Ixxxv to cxxx there is a longitudinal series 
of paired, supra-intestinal glands on the dorsal face of the gut, the 
glands small and in the anterior portions of the segments, in contact 
with the anterior septa and the dorsal blood vessel (the two glands of 
a segment separated from each other by the dorsal trunk). Glands are 
shortly ellipsoidal to almost spheroidal and increase in size slightly 
passing posteriorly into a middle region, from thence decreasing in 
size posteriorly. Anteriorly in a few specimens the glands may be 
flattened and with a triangular dorsal outline, the base and one side 
along the dorsal blood vessel and the anterior septum. On lifting the 
gland slowly and carefully from the gut a short and slender, white 
hollow cord, presumably a duct becomes visible, passing from the 
ventral side of the gland into the gut wall. On pulling the gland still 



140 bulletin: museum of comparative zoology 

further away from the intestine the stalk may break or may come out 
from the gut wall leaving a tiny circular aperture. Glands are located 
as follows: Ixxxvii-cxxiv (1), xci-cxxii (1), xcii-cxx (1), xci-cxxxii (1), 
xcv-cxxix (1). Just behind each gland the transverse blood vessel on 
the dorsal face of the gut may have around it in irregular masses a 
slightly yellowish material (chloragogen?). 

The dorsal blood vessel (single) is continued anteriorly into vii (13) 
where it usually terminates abruptly with the hearts of that segment. 
In several specimens a tiny stub projects anteriorly beyond the com- 
missures or a very small vessel continues the trunk onto the dorsal face 
of the oesophagus from whence it is impossible to trace the vessel an- 
teriorly. A median supra-oesophageal trunk is replaced by a pair of 
longitudinal vessels on the oesophagus in vii-xii, in contact mesially 
in the regions of septal attachment (and united?). These vessels are 
small, usually empty except at or near to junctions with other vessels 
and adherent to the gut (from which they can be dissected off) and 
for these reasons are difficult to trace, unrecognizable posterior to 
12/13 or anterior to 6/7. The ventral trunk is continued anteriorly 
to the region of the subpharyngeal ganglia where it bifurcates, the two 
branches passing laterally. The extra-oesophageal trunk is first recog- 
nizable on each side as a fairly large vessel, just behind and parallel to 
the circumpharyngeal nervous commissures, that passes posteriorly 
close to the ventral trunk and at about the same level, receiving a large 
vessel from the parietes in iv and from the anterior face of the septum 
in V. Just in front of 5/6 or 4/5 there is a transverse connective be- 
tween the two trunks. In vi the trunks rise to a level above that of the 
ventral vessel but still below the ventral face of the gut and are con- 
tinued posteriorly into ix where they usually decrease in size and be- 
come empty and unrecognizable. In one specimen the trunks are 
traceable through x below the median calciferous gland to which 
branches are given and possibly continued thence into xi to the ventral 
face of the posterior gland. In one specimen there are clearly visible 
two longitudinal vessels on the ventral face of the gut in x and xi, 
apparently with connections anteriorly to the extra-oesophageal 
trunks. The subneural trunk is large, behind the prostatic region 
larger than the ventral vessel and looped laterally on both sides of the 
nerve cord, continued anteriorly to a region in front of the subpharyn- 
geal ganglia. Connectives with the extra-oesophageal or other longi- 
tudinal trunks are unrecognizable. 

In xiii a large branch from the dorsal trunk passes ventrally on each 
side along the posterior face of 12/13 just behind the calciferous gland 



gates: peregrine earthworms 141 

to which it apparently gives off branches. In xii, just behind 11/12, 
a similar large branch from the dorsal trunk on each side passes pos- 
teroventrally to the anterior margin of the calciferous gland and after 
giving off a branch which penetrates into an upper portion of the gland 
passes ventrally along the anterior or anteromedian aspect of the 
gland. In vii-xi there are large, heart-like commissures, the last hearts 
always in xi (13). In viii-xi each heart bifurcates dorsally, the pos- 
terior branch passing into the dorsal trunk, the anterior bifurcation 
passing towards the dorsal face of the gut. In x and xi the anterior 
bifurcation joins a vertical vessel from the dorsal face of the calciferous 
gland and then opens into a supra-oesophageal or else passes directly 
into the supra-oesophageal close to the junction of the latter with the 
vertical vessel from the gland. In ix-viii the anterior bifurcation 
passes to a supra-oesophageal trunk. Either anterior or posterior 
bifurcations of the hearts of x-xi may contain blood but in ix-x blood 
is present only in the posterior bifurcations, the anterior bifurcations 
slender, white filaments (functional?). Hearts and commissures of 
vii-xi all pass into the ventral trunk. 

Nephridia are on the parietes close to the anterior septum of a seg- 
ment and extend laterally well towards mid dd. Nephrostomes are 
small, on the anterior faces of the septa close to the ventral parietes 
and nearer to the nerve cord than to the a line. Nephridia are usually 
present in xviii. In pinned out specimens nephridia of iii-vii, viii or ix 
are vertically placed on anterior faces of septa rather than transversely 
on the parietes as posteriorly. 

Testis sacs are unpaired, suboesophageal and transversely placed, 
in the posterior portions of x-xi. From each sac there are protuberant 
four distinct lobes. Two smaller lobes from the anterior margin pass to 
the posterior faces of the septa in front and presumably contain the 
testes. Two protuberances, usually much larger, from the lateral mar- 
gin reach up more or less conspicuously at the sides of the gut. Male 
funnels are small, with smooth lips, without spermatozoal iridescence 
(13) and seated on the ental ends of shortly ellipsoidal bodies which are 
thin-walled vesicular enlargements of the deferent ducts, the vesicles 
filled with a closely packed iridescent material. Testicular coagulum 
extends from the main portion of the sac into the four lobes, surround- 
ing the testes, funnels and vesicles. The ventral blood vessel is adherent 
to the floor of each testis sac, the hearts entering through the median 
faces of the lateral protuberances. Seminal vesicles, two pairs in xi and 
xii, are large — even in aclitellate specimens, soft, folded back on them- 
selves or passing above the dorsal blood vessel into the other side of 



142 bulletin: museum of comparative zoology 

their segments. Copulatory chambers are rather bee-hive-shaped and 
conspicuously protuberant into the coelomic cavity of xvii, often to a 
height of two mm. or sHghtly more, occasionally bent over mesially or 
anteroraesially. Prostates are five to eight mm. long and about one mm. 
thick, the duct muscular but slender, about one mm. long, passing into 
the center of the dorsal face of the copulatory chamber. Deferent ducts 
pass lateral to the copulatory chambers and usually into the ectal 
third of the prostate but in one specimen pass into both prostates at a 
point midway between the ectal and ental ends. Ental limbs of the 
Y-gland are adherent to each other, and are slightly unequal in length, 
the longer about one mm. The duct is slender but muscular, about 
two mm. long and passes into the median face of the copulatory cham- 
ber slightly above the ventral parietes. 

Spermathecae are six to eight mm. long, in xiii and xiv. At a point 
about two mm. or slightly more from the ectal end of each spermatheca 
there is attached a diverticulum, comprising a spirally wound, muscular 
stalk three to five mm. long and a soft terminal chamber with an acinous 
appearance. Sessile on the other side of the duct and about opposite 
the diverticular junction there is a nearly spheroidal body. Opening 
into the duct just above the sessile body and on the same side is a 
slender, thin-walled tube which passes to the posterior face of 12/13 
where it turns mesially and is slightly enlarged. The vesicular enlarge- 
ment which can be dissected off from the septum without especial 
difficulty contains the ovary. Ental to the diverticulum the sper- 
mathecal duct gradually becomes less muscular and eventually is 
nearly as thin-walled as the ampulla. The latter is about three mm. 
long, ovoidal to ellipsoidal, fairly sharply marked off from the duct and 
(in clitellate specimens) filled with a fairly closely packed material in 
which there is no iridescence. 

Abnormalities. One specimen has an extra pair of spermathecal 
pores on xv, all of the pores just median to c. There is an extra pair of 
spermathecae in xv, the ducts about three mm. long, the left ampulla 
shrivelled, the right ampulla about one mm. long, vesicular and dis- 
tended with the same sort of material as in the anterior spermathecae. 

Another specimen has a metameric abnormality of the spiral type 
beginning behind 14/15, a spermathecal pore on the right side only, 
two male pores on the right side on xix and xxi in addition to a single 
normal pore on the left side. The left Y-gland is simple, lacking the 
shorter ental limb. The extra copulatory chamber in xxi is of the usual 
height but has no prostate, the male deferent duct of the right side 
opening into the anterior prostate. Associated with the posterior 



gates: peregrine earthworms 143 

copulatory chamber is a Y-gland slightly larger than usual and with 
its duct passing into the center of the dorsal face of the copulatory 
chamber. Within the chamber is a protuberance just over one mm. 
long that looks much more like a penis than a Y-gland porophore 
though of icourse shorter as well as thicker than a normal penis. 
There is no groove. The lumen is very small and slit-like. 

Regeneration. Three specimens have tail regenerates, two, nine, and 
twelve mm. long. In one of these worms the tail was lost in the middle 
of the supra-intestinal gland region. No supra-intestinal glands are 
recognizable on the gut Of the regenerated tail. 

Remarks. Characteristics of the testis sacs, here as in other species, 
are difficult to determine in dissections under water. If the dissection 
is allowed to dry for a sufficient time and then studied in that condition 
it is possible to open the sac and remove the testicular coagulum lea\- 
ing the major portion of the sac wall intact so that the shape can be 
accurately determined as well as the relationships of the various parts. 

One specimen in a late stage of postsexual clitellar regression (as 
indicated by the discoloration of the clitellar segments) has a large 
number of brown discs scattered through the coelomic cavities of vi- 
xxxi ; one disc two and a half mm. long, Ki discs about one mm. long, and 
a much larger number of small discs. 

Diagnosis. Apertures of copulatory chambers transverse slits with 
centers on or lateral to h and reaching mesially to a, just in front of 
17/18. Spermathecal apertures about as Mnde as cd, with centers on 
or median to c. C'litellum saddle-shaped, on xiv-xviii. Setae; he ea = 
3^2"'''. (l(l>}/2^\ ventral setae lacking on xvii. Nephropores begin on 
iv, just lateral to c. Pigmentation red. Length 90-130 mm. Diameter 
(i mm. 

Gizzard in v; intestinal origin close to 14/15; supra-intestinal glands 
in region of Ixxxv-cxxx. Last hearts in xi; hearts of viii xi latero- 
oesophageal; dorsal blood vessel terminates with commissures of vii. 
Testis sacs unpaired and ventral, with two small anterior lobes con- 
taining testes and two larger laterodorsal lobes containing male funnels 
and vesicular enlargements of deferent ducts. Copulatory chamber 
conspicuously protul)erant into coelomic cavity and containing a penis 
about four mm. long and a Y-gland porophore, a groove from the 
Y-gland aperture passing dorsally along the porophore anfl then 
ventrally to tip of penis. 

Disfrihiifion. "The whole of the tropical zone." 



144 bulletin: museum of comparative zoology 

BIBLIOGRAPHY 

a. References for Gordiodrilus 

Baldasseroni, V. 

1913. Monit. Zool. Ital. 24. 

Beddard, F. E. 

1892. Ann. Mag. Nat. Hist. (6), 10. 

1894. Proc. Zool. Soc. London, 1894. 
Quart. J. Mic. Sci. 36. 

1895. A Monograph of the Order Oligocha^ta. Oxford. 
1901. Proc. Zool. Soc. London, 1901. 

COGNETTI, L. 

1907. Boll. Mus. Zool. Torino, 22. 551. 
1910. Ann. mus. stor. nat. Genova, (3), 4. 

MiCHAELSEN, W. 

1897. Mitt. Mus. Hamburg, 14. 
1900. Das Tierreich, 10. Berlin. 
1903. Ark. Zool. Stockholm, 1. 

Die geograjjhische Verbreitung der Oligochaeten. Berlin. 
1907. Reise in Ostafrika, Voeltzkow, 2, (46). 
1910. Abh. Nat. Ver. Hamburg, 19, (5). 

1913. Zoologica, Stuttgart, 68. 

1914. Beitr. Keniitn. l^and- u. Siisswasserfauna Deutsch-Siidwestafrikas, 
Hamburg, 1. 

1915. Ergeb. 2 Deutsch. Zentral-Afrika-Exp. 1910-11, 1, Zool. 1. 
1933. Abh. Senckenberg. Ges. 40. 

1936. Rev. Zool. Bot. Afr. 29. 

1937. Bull. Mus. Comp. Zool. Harvard, 79. 

Stephenson J. 

1923. Oligochaeta, in Fauna of British India. London. 
1928. Ann. Mag. Nat. Hist. (10), 1. 

1930. The Oligochaeta. Oxford. 

1931. Proc. Zool. Soc. London, 1931. 

b. Other References 

Gates, G. E. 

1937. The genus Pheretimn in North America. Bull. Mus. Comp. Zool. 
Cambridge, 80, (8). 

1938. Indian earthworms. V. Nellogaster geii. nov., with a note on Indian 
species of W oodwardieUa . Rec. Indian Mus. Calcutta, 26. 

On some American and Oriental earthworms. In press. 

Stephenson, J. 

1924. On some Indian Oligochaeta, with a description of two new genera 
of Ocnerodrilinae. Rec. Indian Mus. Calcutta, 26. 



Bulletin of the Museum of Comparative Zoology 

AT HARVARD COLLEGE 

Vol. LXXXIX, No. 4 



SCIENTIFIC RESULTS OF A FOURTH EXPEDITION 
TO FORESTED AREAS IN EAST AND CENTRAL AFRICA 

I 

MAMMALS 



By Glover M. Allen and Arthur Loveridge 



With Five Plates 



CAMBRIDGE, MASS., U.S.A. 

PRINTED FOR THE MUSEUM 

February, 1942 



M 24/942" 

i i O K A K t 



No. 1. — Scientific Results of a Fourth Expedition to Forested Areas 

in East and Central Africa 

I 

Mammals 
By Glover M. Allen and Arthur Loveridge 



CONTENTS Page 

Introduction 147 

List of species collected 150 

Systematic discussion 154 

Bibliography 213 

INTRODUCTION 

The collection on which the following report is based, was made by 
the junior author while investigating the herpetological fauna of cer- 
tain forested regions in East and Central Africa. The enquiry was 
carried out on behalf of the Museum of Comparative Zoology with a 
fellowship granted by the John Simon Guggenheim Memorial Founda- 
tion of New York. 

As in previous reports, the identification and taxonomic work have 
been done by the senior author whose conclusions appear under the 
heading Discussion. The field notes, contributed by the junior author, 
are included under: Breeding, Diet, Parasites, Enemies, Native names, 
Measurements, etc. 

When measurements are given serially, they are always in the follow- 
ing order: (1) length from snout to anus; (2) length of the tail without 
terminal hairs; (3) length of hind foot without claws; (4) length of ear 
from tip to notch. In the case of bats a fifth measurement is added: 
(5) length of wing from axilla to tip. All dimensions are in millimetres, 
and, unless otherwise stated, it is those of the largest male and largest 
female of the series which are supplied. 

The period of collecting mammals was from November 4, 1938, to 
July 25, 1939, during which time, and exclusive of nearly 200 bat and 
embryo alcoholics, 612 skins and skulls representing IIG species or 
races of mammals were secured. Of these, 40 forms were new to the 



148 bulletin: museum of comparative zoology 

collections of the Museum of Comparative Zoology. A very high per- 
centage of these were topotypes of species described from the Ruwen- 
zori Mountains. As much of this collecting was carried on near the 
frontiers of Uganda and Tanganyika, it has resulted in the addition of 
many species to the known fauna of these two countries. 

It has been found necessary to describe as new the following sub- 
species : 

Mops angolensis orientis from Kitaya, Tanganyika Territory. 
Phataginus tricuspis mabirae from Mabira Forest, Chagwe, Uganda. 
Funisciurus pyrrhopus mctoriae from Kibale Forest, Toro, Uganda. 
Thamnomys tenustus hivuensis from Idjwi Island, Belgian Congo. 
Leggada hnfo nhlutus from Idjwi Island, Belgian Congo. 

Altitudes, and detailed information regarding the localities in which 
collecting was carried on, will be published in the final report of this 
series, but as considerable time is likely to elapse before its publication, 
a map (see pi. 1) is furnished giving their approximate position. How- 
ever, many of the forests are of considerable size; Mabira, for example, 
covers some 120 square miles; so below, in parenthesis, is given the 
name of the actual camping site — unlikely to be found on any map — 
and other information likely to be helpful. 

Uganda: Budongo Forest (Bisu); Bugoye (E. foothills of Ruwen- 
zori) ; Bundibugyo (N.W. foot of Ruwenzori) ; Butiaba (N.E. shore of 
L. Albert); P'ort Portal (Provincial headquarters of Toro); Jinja (N. 
shore of L. Victoria) ; Kibale F'orest (same as the Mpanga Forest of the 
Ruwenzori Expedition 1905-6, but Woosnam crossed the Mpanga 
River about 10 miles north of where Loveridge camped near Isunga); 
Mabira Forest (Mubango) ; Mihunga (Loveridge camped on the actual 
site of the Ruwenzori ^Expedition's camp, called by them Mubuku 
Valley, 6000-7000 ft.); Mubuku Valley (Mobuku River on latest 
Uganda Survey map is admittedly wrong, though map spelling was fol- 
lowed on Loveridge's labels, the other rendering, ha\ ing appeared in 
zoological literature, is adhered to here. In this connection it might 
be pointed out that the type locality of "Ruwenzori East" used by 
Thomas in his earlier papers, is synonymous with Mubuku Valley 
between 6000 and 10,000 feet, as will be seen by consulting the final 
report (1910) in Trans. Zool. Soc. London, 19). In citing these type 
localities, therefore, for "Ruwenzori p]ast" we have substituted Mu- 
buku Valley and the precise altitudes, the lowest of which corresponds 
to Loveridge's 'Mihunga'. Mushongero (Mushungero on labels, but 
here again, as former spelling has been used in zoological papers by 



ALLEN AND LOVERIDGE:: AFRICAN MAMMALS 149 

Pitman, it is followed here. Mushongero is a village on the N.E. shore 
of L. Mutanda, Kigezi); Nyakabande (a rest camp on the Kabale- 
Rutschuru road, Kigezi). 

Tanganyika Territory: Amboni Estate (near Amboni village 
about 15 miles N. of Tanga) ; Kitaya (on Rovuma or Ruvuma River, 
about 20 miles inland) ; Lake Rutamba (or Lutamba, about 20 miles 
S.W. of Lindi) ; Lindi (seaport) ; Magrotto Mountains (20 miles from 
Usambara Mountains; camped on Magrotto Estate at edge of forest); 
Mbanja (10 miles north of Lindi, camped on landing field); Mikindani 
(on hill 3 miles N. of township) ; Nchingidi (in forest on Rondo Plateau, 
about 50 miles S.W. of Lindi) ; Siga Caves (about 10 miles N. of Tanga, 
7iot Sigi at foot of Usambara); Tanga (at hotel, before sailing). 

Kenya Colony: Mainland opposite Kilindini ferry (collected near 
ferry landing). 

Belgian Ruanda : Kisenyi (N. shore of L. Kivu, camped on roadside 
at Kiraga 3 miles N. of Kisenyi). 

Belgian Congo: Idjwi Island (Kwidjwi, camped on upper reaches 
of Mulinga River, circa 4500 ft.). 

A selection of duplicates of such species as were collected in the 
Belgian Congo and Ruanda, are to be sent to the Congo Museum, 
Tervueren, in appreciation of the action of His Excellency the Gov- 
ernor of the Congo Beige in granting a permit to collect during the 
month spent in these countries. 

We also take this opportunity of thanking our colleague Dr. Joseph 
Bequaert for his kindness in identifying the ectoparasites recorded in 
this paper. Dr. H. R. Hill of Los Angeles Museum for naming lin?uatu- 
lids, also Dr. B. Schwarz and Dr. J. T. Lucker of the United States 
Department of Agriculture for similar courtesies regarding helminths, 
and the Rev. Lyndon Harries for eliminating some plural prefixes of 
native names obtained in southeastern Tanganyika. 

For permission to use the blocks of plates 2 and 5, we are in- 
debted to the Editor of the Scientific Monthly in which journal (June 
and July, 1940) they appeared as illustrations to a popular account 
of the safari. 



150 bulletin: museum of comparative zoology 



INDEX TO SPECIES COLLECTED 

MACROSCELIDIDAE Page 

Rhynchocyon petersi melanurus Neumann 154 

Rhynchocyon petersi macrurus Giinther 154 

Petrodromus sultani sultani Thomas 155 

Petrodromvs rovumae Thomas 155 

SORICIDAE 

Sylvisorcx granti granfi Thomas 155 

Myosorcx hlarina Thomas 156 

Crocidura nyansae kicu Osgood 156 

Crocidura hirta kirta Peters 156 

Crocidura hirta velutina Thomas 157 

Crocidura sacralis Peters 158 

Crocidura hUdcgardeae hUdcgardeae Thomas 158 

Crocidura hicolor elongius Osgood 159 

Crocidura hicolor Viendersoni Dollman 159 

Crocidura littoralis Heller 159 

PTEROPIDAE 

Eidolon helvum helvum (Kerr) 160 

Epomophorus wahlhergi wahlbcrgi (Sundevall) 160 

Rousettus angolensis (Bocage) 160 

EMBALLONURIDAE 

Coleura afra (Peters) 161 

Taphozotis mauritianus viauritianv3 E. Geoffroy 161 

NYCTERIDAE 

Nycteris aethiopica oriana Kershaw 161 

Nycteris thebaica subsp 162 

RHINOLOPHIDAE 

Rhinolophus hildchrandtii hildehrandtii Peters 162 

Rhinolojihus fumigatus exsul Andersen 162 

Rhinolophus lobatus Peters 163 

HIPPOSIDERIDAE 

Hippos ideros cy clops (Temminck) 163 

Hipposideros gigas gigas (Wagner) 164 

Triaenops afer Peters 165 



ALLEN AND LOVERIDGE: AFRICAN MAMMALS 151 

VESPERTILIONIDAE ~ Page 

Pipistrellns nanus nanus (Peters) 165 

Mmiopteriis minor Peters 166 

MOLOSSIDAE 

Mops angolcJisis orientis subsp. nov 166 

Chaerephon limbatiis (Peters) 167 

Chaerephon pumilus naivashae Hollister 168 

MUSTELIDAE 

Poecilogale albinucha doggetti Thomas & Schwann 168 

Aonyx capensis Ihindei (Thomas) 169 

Lutra macuUcolUs tenuis Pohle 169 

VIVERRIDAE 

Genetta tigrina stuhlmanni Matsehie 170 

Genetia tigrina suahelica Matsehie . . . . " 170 

Myonax sanguineus protcus (Thomas) 170 

Atilax paludinosus Imordax (Thomas) 171 

Bdeogale crassicauda omnivora Heller 171 

FELIDAE 

Felis hrachyiira panlasticta Pocock 172 

LORISIDAE 

Gal ago crassicaudatus lasiotis Peters 172 

Galago demidovii thomasi Elliot 173 

Galago seiiegalensis zanziharicus Matscliie 173 

Galago senegalensis moholi Smith ^. 173 

CERCOPITHECIDAE 

Cercocebus albigena johnstoni (Lydekker) 174 

Cercopithecus niditans schmidti Matsehie 174 

Cercopithecus Vhoesti Vhoesti Sclater 175 

Cercopithecus mitis stuhlmanni Matsehie 175 

Cercopithecus mitis schoutedeni Schwarz 175 

Cercopithecus mitis monoides Geoffroy 176 

Cercopithecus aethiops centralis Neumann 176 

Cercopithecus aethiops johnstoni Pocock 177 

COLOBIDAE 

Colobus polykomos palliatus Peters 177 

Colobus badius tephroscelcs Elliot 177 



152 bulletin: museum of comparative zoology 

MANIDAE Page 

Phataginus tricuspis mabirae subsp. nov 178 

SCIURIDAE 

Aethosciurus ruwevzorii ruwenzorii (Schwann) 179 

Heliosciurus rufobrachium arrhenii Lonnberg 179 

Heliosciurus mfobrachium ny ansae (Neumann) 179 

Heliosciurus undvlatus undulatus (True) 180 

Funisciurus pyrrhopus victoriae subsp. nov 180 

Paraxcrus palliatus frerei (Gray) 181 

Paraxerus sponsus bridgemani Dollman 182 

Paraxerus ochraceus aruscensis (Pagenstecher) 182 

Paraxerus flavivittis exgeanus Hinton 183 

Tamiscus alexandri (Thomas & Wroughton) 183 

Tamiscus emini emini (Stuhlmann) 184 

Protoxerus stangeri centricola (Thomas) 185 

MUSCARDINIDAE 

Claviglis murinus soleatus (Thomas & Wroughton) 186 

CRICETIDAE 

Tatera cosensi (Kershaw) 186 

Tatera nigrita nigrita Wroughton 187 

Tatera ruwenzorii Thomas & Wroughton 187 

RHIZOMYIDAE 

Tachyoryctes ruandae Lonnberg & Gyldenstolpe 187 

MURIDAE 

Dendromus insigris kivu Thomas 188 

Dendromus messorius ruddi Wroughton 189 

Dendromus whytei pallescens Osgood 189 

Thamnomys surdaster surdasier Thomas & Wroughton 190 

Thamnomys surdaster dryas Thomas 191 

Thavfmomys venustus venustus Thomas 191 

Thamnomys venustus kivuensis subsp. nov 192 

Oenomys kypoxanthus unyori (Thomas) 193 

Oenomys hypoxanthus editus Thomas & Wroughton 193 

Rattus rattus kijabius (Allen) 194 

Aelhomys kaiseri hindei (Thomas) '. 195 

Praomys jacksoni montis (Thomas & Wroughton) 196 

Praomys taitae (Heller) 196 



ALLEN AND LOVERIDGE: AFRICAN MAMMALS 153 

Page 

Hylomyscus denniae denniae (Thomas) 197 

Hylomyscus carillus schoutedeni (Dollman) 197 

Mastomys concha ugandae (De Winton) 198 

Mastomys concha durumae (Heller) 198 

Mastomys concha microdon (Peters) 198 

Lcggada bnfo bufo Thomas 199 

Leggada bvfo abhitns subsp. nov 199 

Leggada grata grata Thomas & Wroughton 200 

Leggada bella vidna Thomas 201 

Cricetomys gambianus proparator Wroughton 202 

Cricetomys gambianus osgoodi Heller 203 

Lophuromys aquilus aquilus (True) 203 

Lophnromys aquilus subsp 203 

Lophuromys aquilus laticeps Thomas 204 

Lophuromys tvoosnami woosnami Thomas 205 

Acomys wilsoni tvilsoni Thomas 205 

Acomys ? albigena Heuglin 206 

Dasymys bentleyae medins Thomas 206 

Pelomys fallax concolor Heller 207 

Pelomys fallax insignatus Osgood 207 

Arvicanthis abyssinicus nubilans Wroughton 208 

Lcmniscomys striatus massaicus (Pagenstecher) 208 

Otomys derdi Thomas 209 

Otomys kempi Dollman 209 

BATHYERGIDAE 

Cryptomys hottentotus whytei (Thomas) 210 

HYSTRICIDAE 

Hystrix africaeanstralis ? subsp 210 

LEPORIDAE 

Lepus capensis ? abbotti HoUister 211 

SUIDAE 

Hylochoern^ meinertzhageni meinertzhageni Thomas 211 

BOVIDAE 

Cephalophus caerulus aequaiorialis Matschie 211 

Sylvicapra grimmia ? roosexelti Heller 212 



154 bulletin: museum of comparative zoology 

Systematic Discussion 

MACROSCELIDIDAE 

Rhynchocyon petersi melanurus Neumann 

Rhynchocyon -petersi melanurus Neumann, 1900, Zool. Jahrb., Syst., 13, p. 542: 
Lindi, Tanganyika Territory. 

& 9 (M. C. Z. 38781-2) Nchingidi, T. T. 20. v. 39. 

Discussion. Nchingidi, on the Rondo Plateau, lies about fifty miles 
by road and trail southwest of Lindi. These specimens, therefore, are 
nearly topotypes, and agree with Neumann's brief diagnosis. The nape 
and sides are of a rich chestnut, the lower back is much darkened with 
black. In both, however, may be made out in certain lights the two 
longitudinal black stripes lying one on each side of the mid-dorsal line, 
but in one the outer border of each stripe has plainly the series of in- 
dentations which in the race macrurus partly enclose squarish black 
spots. The tails have nearly the terminal third white all round, with a 
small black tip, as mentioned by the describer. 

Rhynchocyon petersi macrurus Giinther 

Rhynchocyon macrurus Giinther, 1881, Proc. Zool. Soc. London, p. 163: 
Rovuma River, 8° S., Tanganyika Territory. 

cT (M. C. Z. 38783) Kitaya, Rovuma River, T. T. 4. iv. 39. 

Native names. Litotwe (Kiyao); norda (Kimakonde). 

Discussion. A single immature specimen from Kitaya may be taken 
as typical of this ^elephant-shrew, which is obviously a form of R. 
petersi, and represents more nearly the most primitive condition of the 
color pattern. The head forward from the crown is a mixture of ochra- 
ceous and black; the nape is chestnut with a narrow black median line; 
the flanks, hips, and belly are chestnut. On the back the pattern of 
stripes is evident, and unobscured by black. On each side of the mid- 
line is a series of squarish chestnut spots, connected on the median side 
by a slightly darker stripe, the two of opposite sides meeting at the base 
of the tail. External to this stripe are two more, the upper consisting 
of some half a dozen alternating chestnut and ochraceous spots, the 
lower (outer) of smaller and slightly paler spots. The space between 
the side stripes is a mixed ochraceous and black like the forehead. The 
pattern thus somewhat resembles that of the checker-backed shrews, 
but the ground color is chestnut instead of grayish. Evidently the 



ALLEN AND LOVERIDGE: AFRICAN MAMMALS 155 

typical R. petersi is a form in which the original checkered pattern is 
completely obscured by the black back. 

Measurements, cf juv. 183. 170. 60. 22 mm. 

Petrodromus sultani sultani Thomas 

Petrodromus sultan Thomas, 1897, Proc. Zool. Soc. London, p. 435: Mombasa, 
Kenya Colony (misprint of specific name corrected 1898). 

9 (M. C. Z. 38788) Amboni Estate, near Tanga, T. T. 21. vi. 39. 

Measurements. 9 . 200. 160. 53. 34 mm. 

Diet. Termites only in the stomach so far as could be seen. 

Remarks. With my spotlight focussed on its large red eyes, I saw- 
one of these elephant shrews repeatedly kicking its left hind leg on the 
dry leaves, apparently a nervous action of alarm like the stamping of a 
rabbit. 

Petrodromus rovuivlve Thomas 

Petrodrom.us rovumae Thomas, 1897, Proc. Zool. Soc. London, p. 434: Rovuma 
River, 100 miles inland, Tanganyika Territory. 

9 (M. C. Z. 38785) Mbanja, T. T. 6. v. 39. 
2 c^ 1 9 (M. C. Z. 38784, 38786-7) Nchingidi, T. T. 16-20. v. 39. 

Native name. Ntotwe (Kimakonde). 

Discussion. The females are much buffier below, as HoUister (1918, 
p. 29) has noted. A male (M. C. Z. 38784) has a few knobbed bristles 
on the underside of the tail. 

Measurements, d^. 185. 187. 51. 44 mm., 9 • 160. 161. 52. 38 mm. 

Remarks. The Wakonde at Mbanja employed a bag net to capture 
this giant shrew. 

SORICIDAE 

Sylvisorex granti granti Thomas 

Sylvisorex granti Thomas, 1907, Ann. Mag. Nat. Hist. (7), 19, p. 118: Mubuku 
VaUey, Ruwenzori Mountains, Uganda, 10,000 ft. 

1 ?, 1 9,2 yng. (M. C. Z. 39307-10) Mihunga, U. 28. .xii. 38 & 14. 
i. 39. 

Native name. Mususu (Lukonjo). 

Measurements. ? sex. 66. 52. 11. 7 mm., 9 . 63. 57. 11. 6 mm., young 
40. 31. 9. 5 mm. 

Breeding. As we were engaged in demolishing the stump of a wild 
banana, already wrecked by elephant, at the edge of a swamp in the 



156 bulletin: museum of comparative zoology 

Mubuku Valley, but at 6,500 feet, the female shrew appeared running 
round the base of the plant while pulling two young after her. Each 
of these young was attached by one of the four nipples situated very 
far back between the 'hind limbs (14.i.39). 

Mtosorex blarina Thomas 

Myosorex blarina Thomas, 1906, Ann. Mag. Nat. Hist. (7), 18, p. 139: Mubuku 
Valley, Ruwenzori Mountains, Uganda, 10,000 ft. 

d" (M. C. Z. 39269) Idjwi Island, B. C. 2. iii. 39. 

Native name. Mushushu (Lulega). 

Discussion. This single skin from an island in Lake Kivu agrees in 
its blackish color and short tail with Thomas and Wroughton's figure, 
and is an interesting record for a species hitherto known from the 
Ruwenzori Mountains. 

Measurements, cf . 67. 40. 14. 9 mm. 

Crocidura nyansae kivu Osgood 

Crocidura flavescens kivu Osgood, 1910, Ann. Mag. Nat. Hist. (8), 5, p. 370: 
Lake Kivu, Belgian Congo. 

9 (M. C. Z. 39198) Bugoye, U. 23. i. 39. 

Discussion. This race, slightly smaller and more richly colored than 
the typical form, has previously been reported from Ruwenzori at 
altitudes of from 5000 to 7000 feet. 

Measurements. 9 . 130. 74. 18. 10 mm. 

Remarks. Taken at the rest camp in the Ruwenzori foothills, in a 
rat trap baited with bread! 

Crocidura hirta hirta Peters 

Crocidura hirta Peters, 1852, Beise nach Mossambique, Siiugeth., p. 78, pi. 
xviii, fig. 2: Tette, Mozambique. 

1 9 (M. C. Z. 38796) Kitaya, Bovuma River, T. T. 28. iii. 39. 
5 o^ 7 9 (M. C. Z. 38792-5, -98-801, -819-22) Mikindani, T. T. 

11-20. iv. 39. 
1 d' 1 9 (M. C. Z. 38789, 38797) Mbanja, near Lindi, T. T. 27. 
iv-11. V. 39. 
3 9 (M. C. Z. 38790-1, 38796) Nchingidi, Rondo, T. T. 17-19. 

V. 39. 
1 cf (M. C. Z. 39064) Lindi township, T. T. 1. vi. 39. 



ALLEN AND LOVERIDGE: AFRICAN MAMMALS 157 

Native names. Chanyunga (Kimwera); nanyunga (Kimakonde at 
Mikindani); ntawara (Kimakonde at Mbanja). 

Discussion. In this series from southeastern Tanganyika, the second 
and third small unicuspids are practically equal in size in both crown 
area and profile view. The skins show the variations described by 
Dollman (1915, p. 71): they are grayer in immature animals, but cin- 
namon or chocolate brown in those with pelage fully grown, variations 
formerly believed to indicate different species. The lateral gland in 
both sexes, when adult, is usually marked by a lengthwise spot of 
appressed hairs. 

Measurements. Largest cT. 110. 64. 15. 11 mm., 9 . 100. 62. 15. 11 
mm. Both from Mikindani. 

Breeding. On March 28, at Kitaya, an 85 mm. female and her single 
newborn naked young found lying on damp earth beneath a pile of 
weeds at the edge of a rice swamp; a surprisingly damp situation but 
teeming with insect life. On May 1, at Mbanja, a nest containing a 
single naked nestling. On May 3, another nest held four young, the 
largest of which measured 55. 13. 10. 5 mm. (alcoholics). 

Enemies. One recovered from the stomach of a cobra {Naja n. nigri- 
coUis) and a young one from a burrowing viper {Atractaspis hihronii), 
both at Mbanja. 

Crocidura hirta velutina Thomas 

Crocidura velutina Thomas, 1904, Ann. Mag. Nat. Hist. (7), 14, p. 237: Usam- 
bara Mountains, Tanganyika Territory. 

11 d^ 10 9 (M. C. Z. 39066-8, 39078-95) Amboni Estate, T. T. 
19-21. vi. 39. 
9 (M. C. Z. 39065) Magrotto Mountain, T. T. 1. vii. 39. 

Native name. Keke (Kisambara). 

Discussion. Although Dollman (1915, p. 79) in his review of the 
genus, gives C. velutina the rank of a distinct species, these specimens 
from northeastern Tanganyika are so closely related to both C. hirta 
and C. hindei, there seems no doubt that it is actually but a slightly 
marked race of the former, with which it agrees in size and color, but 
differs in that the third upper unicuspid is usually distinctly smaller 
than the second in side view. We therefore give it doubtful subspecific 
standing. In one individual (M. C. Z. 39092) it is missing on the left 
side. Magrotto Mountain is but twenty miles from the type locality, 
Amboni village about fifty. 



158 bulletin: museum of comparative zoology 

Measurements. Largest cT. 110. 60. 15. 11 mm., $ . 101. 56. 14. 11 
mm. Both from Amboni. 

Habitat. The entire Amboni series were living in piles of vegetable 
debris in a sisal plantation, and were captured when a tractor was en- 
gaged in demolishing the piles and spreading them over the ground. 
The Magrotto shrew was sleeping beneath a bundle of sedges in a 
swamp. 

Crocidura sacralis Peters 

Crocidura sacralis Peters, 1852, Reise nach Mossambique, Saugeth., p. 82, pi. 
xviii, fig. 3: Cabageira Peninsula, Mozambique. 

1 cf 2 9 (M. C. Z. 38921-2, 39060) Lindi, T. T. 31. v. 39. 

Discussion. The three specimens agree closely with the description 
of this species originally named from Mozambique. The grayish- 
cinnamon back, and grayish-white belly are much paler than the 
coloration of C. hildcgardeae; the tail is slightly shorter than the head 
and body; the skull is longer (21 mm.) and the large anterior cusp on 
the last upper premolar is much reduced so that it is hardly noticeable 
as a low cingulum. 

Measurements. Both & and 9 9 . 75. 40. 14. 5 mm. 

Habitat. All taken under bundles of thatching grass in native town. 

Crocidura hildegardeae hildegardeae Thomas 

Crocidura hildegardeae Thomas, 1904, Ann. Mag. Nat. Hist. (7), 14, p. 240: 
Fort Hall, Kenya Colony. 

1 (^ 3 9 (M. C. Z. 40739-42) Kisenyi, B. R. 10. ii. 39. 

2 cf 2 9 (M. C. Z. 40735-8) Idjwi Id., B. C. 21. ii & 2-3. iii. 39. 

Native name. Mushushu (Lulega). 

Discussion. The small size, gray belly, uniformly dark tail, small 
foot (13 mm. with claws), the skull of 19-20 mm. total length, upper 
tooth row 8-8.5 mm., distinguish this species, which is widespread in 
East Africa. In their dull, dark-brown color these specimens agree 
with others from Kenya Colony and Tanganyika Territory. 

Hollister (1918, p. 64) has shown that the various described forms 
from eastern Africa are possibly unrecognizable as distinct races and 
that very probably the name gracilipes Peters should replace hilde- 
gardeae of Thomas. 

Measurements. Largest cf . 70. 50. 13. 8 mm., 9 . 80. 49. 13. 8 mm. 
Both from Idjwi Id. 



ALLEN AND LOVERIDGE: AFRICAN MAMMALS 159 

Breeding. The Kisenyi series, all taken in one nest, evidently con- 
sist of a mother ( 9 . 65. 48. 12. 6 mm.) and her family, of which the 
male measured 57. 45. 12. 6 mm. 

Enemies. One recovered from the stomach of a sedge viper {Athens 
nitschei). 

Crocidura bicolor elgonius Osgood 

Crocidura bicolor elgonius Osgood, 1910, Ann. Mag. Nat. Hist. (8), 5, p. 369: 
Kirui, Mt. Elgon, Kenya Colony. 

9 (M. C. Z. 39207) Butiaba, U. 5. xii. 38. 

Discussion. This specimen is of the same slaty brown as one from 
Kaimosi, Kenya Colony, and has a skull length of 16 mm. 
Measurements. 9 . 65. 46. 9. 7 mm. 
Habitat. Under rotting vegetation at the edge of a swamp. 

Crocidura bicolor ?hendersoni Dollman 

Crocidura bicolor hendersoni Dollman, 1915, Ann. Mag. Nat. Hist. (8), 15, p. 
517: Nj'asaland. 

1 rf' 2 9 (M. C. Z. 38802-^) Mbanja, T. T. 27. iv. 39. 

Native names. Ntatvara (Kimakonde at Mbanja). 

Discussion. In color these three skins are barely distinguishable 
from the Butiaba specimen in being less slaty, but distinctly cinnamon 
brown above, with whitish instead of dusky feet. The skulls measure 
17 mm. in length; front of incisor to back of large premolar, 5 mm.; 
hind foot small, 8 mm. They may represent the race hendersoni. 

Measurements, cf . 52. 40. 8. 8 mm., 9 . 58. 35. 8. 8 mm. 

Habitat. All taken together, according to the native who brought 
them in. 

Crocidura littoralis Heller 

Crocidura littoralis Heller, 1910, Smithsonian Misc. Coll., 56, No. 15, p. 5: 
Butiaba, east shore Albert Nyanza, Uganda. 

2 9 (M. C. Z. 39270-1) Idjwi Id., B. C. 28. ii. & 3. iii. 39. 

Native name. Mushushu (Lulega). 

Discussion. These two shrews agree with Heller's description in 
their dark slaty color and faint brownish cast, as well as in having but 
few scattered bristle hairs on the tail. The skulls agree in their larger 
size as compared with niobe of Ruwenzori, having a total length of 
24 mm. in the adult with a tooth row of 10.5 against 20 and 8.2, re- 



160 bulletin: museum of comparative zoology 

spectively. This record involves a considerable southward extension 
of the known range. 

Measurements. 9 9 . 88, 82 mm., 65, 61 mm., 17, 15 mm., 10, 5 mm. 

PTEROPIDAE 

Eidolon helvum helvum (Kerr) 

Vespertilio vampyrus helms Kerr, 1792, Linnaeus's Animal Kingdom, 1, pt. 1, 
pp. .xvii, 91 : No locality. 

9 (M. C. Z. 39197) Mihunga, U. 12. i. 39. 

Native navies. Kakorokombi (Lukonjo), chugugu (Lutoro). 

Breeding. Held a large fetus whose head measured 35 mm. 

Habitat. Shot from a group of about fifty which we disturbed in a 
clump of dracaena in the ravine immediately below our camp on 
Ruwenzori. On returning to the (palms) the others clambered up 
among the tangle of drooping, withered leaves. 

Epomophorus wahlbergi wahlbergi (Sundevall) 

Pteropus imhlbergi Sundevall, 1846, Ofversigt af Kongl. Svenska Vet.-Akad- 
Forhandl., 3, No. 4, p. 118: Near Fort Natal and in interior of Caffraria. 

9 (M. C. Z. 38840) Mikindani, T. T. 15. iv. 39. 

Native names. N'ncma (Kimakonde), nema (Kimwera). 

Measurements. 9 . 123. 0. 20. 25. 280 mm. 

Remarks. Taken in the bat net together with a Triaenops afer; the 
latter was not noticed when at daybreak I transferred the section of 
the net holding the fruit bat to a cyanide tin. On removing it half 
an hour later, the fruit bat was found to be dead but much bitten 
about the breast by the afer, which was still alive. It seemed a strange 
coincidence that in a net of 60 feet by 8 feet the only two bats captured 
should be taken at the same spot! 

Rousettus angolensis (Bocage) 

Cynonycteris angolensis Bocage, 1898, Jorn. Sci. Lisboa, (2), 5, pp. 133, 138, 
fig.: Pungo Andongo; Cahata; Quibula, Angola. 

1 d^ 8 9 (M. C. Z. 38963-71) Magrotto Mtn., T. T. 11. vii. 39. 

Native name. Ndema (Kisambara, but not specific). 

Discussion. This bat is a member of the subgenus Lissonyctcris. 
The short tibia, about 30 mm., short tail, and wing attached to the 
back of the second toe, readily distinguish this species from R. leachi. 



ALLEN AND LOVERIDGE: AFRICAN MAMMALS 161 

The skull length of M. C. Z. 38963 is only 39 mm., or slightly smaller 
than that given by Andersen (42. 5-44). 

Measurements, cf . 125. 20. 21. 23. 261 mm., 9 . 137. 20. 21. 23. 265 
mm. 

Parasites. All were swarming with nycteribiids ( — ). 

Habitat. One was netted at the forest edge, all the rest were taken 
at the entrance of Kitulwe Cave, higher up the mountain. 

EMBALLONURIDAE 

CoLEURA AFRA (Peters) 

Emhallonura afra Peters, 1852, Reise nach Mossambique, Saugeth., p. 51, pi. 
xii, pi. xiii, figs. 18-19; Tette, Mozambique. 

5 c? 5 9 (M. C. Z. 38923,-47-53,-90-91) Siga Caves, T. T. 8. vi. 39. 
19 cf 3 9 (alcoholics) Siga Caves near Tanga, T. T. 8. vi. 39. 

Measurements, d". 63. 17. 9. 18. 145 mm., 9 . 64. 18. 9. 16. 150 mm. 

Habitat. All netted at the entrance to one of the smaller caves. 

Parasites. An arachnid-like dipteron ( ) 

on one. 

Enemies. One recovered from the stomach of a bat hawk (Machaer- 
hamphus anderssoni). 

Taphozous mauritl^.nus mauritianus Geoffroy 

Taphozous mauritianus E. Geoffroy, 1818, Description de I'Egypte, 2, p. 127: 
Mauritius. 

1 d'2 9 (M! C. Z. 38837-9) Mbanja, T. T. 28. iv. 39. 

Native name. Kiputiputi (Kimakonde, but not even generic). 
Measurements. cT. 85. 24. 11. 16. 190 mm., 9 . 87. 24. 13. 18. 190 mm. 

Parasite. A nycteribiid ( ). 

Habitat. Abundant on coconut palms, living in pairs. 

NYCTERIDAE 

Nycteris aethiopica oriana Kershaw 

Nycteris oriana Kershaw, 1922, Ann. Mag. Nat. Hist. (9), 10, p. 179: Chiromo, 
Shire Valley, Nyasaland. 

9 (M. C. Z. 38818) Mbanja, T. T. 27. iv. 39. 

Native name. Kiputiputi (Kimakonde, but not even generic). 



162 bulletin: museum of comparative zoology 

Discussion. This specimen agrees fairly well with the description of 
this race: tragus crescentic, forearm 51 mm., tibia and extended foot 
35 mm., skull length 22 mm. The record involves an eastward exten- 
sion of the range. 

Measurements. 9 . 61. 60. 11. 30. 160 mm. 

Nycteris thebaica subsp. 
12 cf 1 9 (M. C. Z. 38805-17) Mbanja, T. T. 4. v. 39. 

Discussion. This series includes two full-grown but dark immature 
individuals. All agree in cranial characters and measurements with 
N. thebaica and possibly represent the race aurantiaca de Beaux, de- 
scribed from the Northern Guaso Nyiro, Kenya Colony. They are 
drabby brown above, and dull grayish below with a clearer ochraceous- 
tawny tint on the sides of the neck. 

Measurements. cT. 55. 51. 8. 35. 133 mm., 9 • 55. 51. 8. 34. 141 mm. 



RHINOLOPHIDAE 

Rhinolophus hildebrandtii hildebrandtii Peters 

Rhinolophus hildebrandtii Feters, 1878, Monatsb. Akad. Wiss. Berlin, p. 195, 
pi. i, figs. 1-la: Ndi, Teita, Kenya Colony. 

d^ (M. C. Z. 38823) Mbanja, T. T. 4. v. 39. 

c^ (M. C. Z. 38982) Magrotto Mtn., T. T. 11. vii. 39. 

Discussion. As pointed out by Hollister (1918, p. 84), the larger 
size readily distinguishes this species from R. eloquens (forearm 56 mm.) 
which it in general resembles. 

Measurements, cf . 112. 44. 13. 39. 185 mm., from Mbanja. 

Habitat. The Mbanja bat, together with the series of Nycteris 
thebaica listed above, was captured by netting the doorway of a small 
basement room in the ruins of Chief Masudi's father's home on a 
mangrove-grown estuary. 

Rhinolophus fumigatus exsul K. Andersen 

Rhinolophus fumigatus exsid K. Andersen, 1905, Ann. Mag. Nat. Hist. (7), 16, 
p. 64: Kitui, Kenya Colony. 

cf (M. C. Z. 38824) Mbanja, T. T. 3. v. .39. 

Discussion. In the upper jaw the anterior premolar is minute, 
crowded to the outer side of the tooth row and barely reaching the 



ALLEN AND LOVERIDGE: AFRICAN MAMMALS 163 

cingulum level of the posterior premolar, which is in contact with the 
canine. In the lower jaw the corresponding tooth is absent. The fore- 
arm measures 50 mm. 

Measurements, cf . 58. 26. 9. 24. 150 mm. 



Rhinolophus lobatus Peters 

Rhinolophus lobatus Peters, 1852, Raise nach Mossambique, Siiugeth., p. 41, 
pi. ix, pi. xiii, figs. 16-17: Sena and Tette, Mozambique. 

6 c? 4 9 (M. C. Z. 38972-81) Magrotto Mtn., T. T. 11. vii. 39. 

Native name. Ndema (Kisambara), 

Discussion . These agree closely with Peters's description and figures, 
though in their pale-based fur with brown stippling these delicately 
formed little bats bear at first sight a resemblance to ITipposideros 
caffer. Hollister (1918, p. 84) has recorded the species from Naivasha 
and Kijabe, Kenya Colony, so that it evidently has a wide range in 
eastern Africa. 

Measurements, cf . 57. 25. 8. 17. 136 mm., 9 . 59. 24. 8. 17. 137 mm. 

Parasite. A large nycteribiid ( ) on one. 

Habitat. Netted at the entrance to Kitulwe Cave near the summit 
of the mountain. 



HIPPOSIDERIDAE 

HiPPOSiDEROS CYCLOPS (Temminck) 

Pyllorrhina cyclops Temminck, 1853, Esquisses Zool. sur la Cote de Guine, 
p. 75: Boutry River, Gold Coast. 

9 (M. C. Z. 40829) Budongo Forest, U. 1. xii. 38. 

Amative name. Kinyira (Luganda), usually reserved for fruit bats. 

Discussion. The specimen agrees closely with one from Avakubi, 
Belgian Congo, a topotype of H. langi, which Hayman has shown to be 
synonymous with H. cyclops. Its forearm (female) is 71 mm., or 
slightly larger than that of a male specimen from Liberia (66.5 mm.), 
but the difference may be merely individual. 

Measurements. 9 . 95. 40. 18. 35. 210 mm. 

Habitat. Netted in the forestry nursery at Bisu, a clearing in the 
heart of the forest. 



164 bulletin: museum of comparative zoology 

HiPPOSiDEROS GiGAS GiGAS (Wagner) 

Rhinolophus gigas Wagner, 1845, Arch. Naturg., 11, p. 148: Benguela, Angola. 

8 d' 2 .9 1 ? (M. C. Z. 38917-20,-34-40) Siga Caves, Tanga, T. T. 
8. vi. 39. 

Native name. Ndema (Kisambara, but not generic). 

Discussion. The discovery of these large bats in northeastern 
Tanganyika is interesting. Andersen in 1906, when reviewing this 
group, stated that specimens had been examined from Angola (the 
type locality) to Gambia (whence he described the race gamhiensis), 
but the only evidence of its occurrence on the eastern side of the con- 
tinent was that afforded by Peters's record (as Phyllorhina vittata) from 
Querimba Islands, Mozambique. Since then, J. A. Allen described in 
1917 a new dark race niangarae, based on a single female from Nian- 
gara, Uele district, northeastern Belgian Congo. The present series 
secured by Loveridge therefore still further extends the known range. 
The specimens in measurements fall within the limits given by Ander- 
sen, with forearms in males 100-110 mm.; upper tooth row 13.7-14.5 
mm.; antorbital width 11-12 mm. In the absence of adequate com- 
parative material, they are regarded as of the typical race. 

Measurements. &. 140. 43. 25. 35. 323 mm., 9 . 120. 45. 23. 31. 325 
mm. 

Habitat. The Siga (or Mkulumu.si) Caves on the banks of the Mkul- 
umusi River, appear to have been eroded from the limestone by the 
former course of the river. They are said to extend for a mile; in places 
the vaulted roof reaches to a great height. Shooting these bats was 
a most eerie experience; no sooner did we enter than there was a noise 
as of rushing water, resulting from the wing beats of thousands of bats 
which had been disturbed by the light from our electric torches. In an 
outer chamber we came upon the smaller horseshoe bats (C afra), then 
a huge Hipposideros passed me so I followed a narrow passage, from 
which it seemed to have come, until we came to a high vaulted chamber 
that was apparently their headquarters. Most of the floor was under 
water. Squeakings in all sorts of keys seemed to indicate many species, 
but only gigas was revealed by my headlight or fell to my shots. At 
each report hundreds of bats came whirling about us while clouds of 
dust, dislodged by the explosion, set us coughing. The floor, carpeted 
by the accumulated bat guano of centuries, was rendered slippery by 
the continual dripping of water from the roof. As the bats fell into the 
pool we recovered them with the aid of a butterfly net. It was inter- 
esting to note that on our return to the cave on the following day not a 



ALLEN AND LOVERIDGE: AFRICAN MAMMALS 165 

single giant horseshoe bat was to be seen — they had all migrated to 
some fresh retreat where they might be free from molestation. 

Trl^enops afer Peters 

Triaenops afer Peters, 1877 (1876), Monatsb. Akad. Wiss. Berlin, p. 913, fig. 2: 
Mombasa, Kenya Colony. 

cf (M. C. Z. 38836) Mikindani, T. T. 15. iv. 39. 
6 d' 6 9 (M. C. Z. 38941-5,-83-9) Siga Caves, T. T. 8. vi. 39. 
19 cf 18 9 (alcoholics) Siga Caves, near Tanga, T. T. 8. vi. 39. 

Native name. Namutimuti (Kimakonde and Kimawiha). 

Discussion. The short ear with its abruptly narrowed tip is char- 
acteristic; forearm 53 mm.; tail extending about two-thirds the dis- 
tance to the edge of the interfemoral membrane. Most of the skins are 
a warm golden brown (about Front's brown of Ridgway) above, some- 
what paler below. Two, however, are uniformly bright cinnamon 
above to cinnamon buff below. Another is evidently molting and re- 
tains a collar of 'orange-cinnamon' as w^ell as a small patch of the same 
in the middle of the back; warm buff below. This bright coloring 
may mark the fully adult stage. Two immature specimens are darker 
gray above, whitish below. 

Measurements, d'. 68. 29. 8. 13. 151 mm., 9 . 65. 30. 10. 15. 145 mm. 

Remarks. The Mikindani male whose measurements are given above, 
was netted with Epomophorus w. ivahlbergi as already related; the Siga 
series were netted at the entrance of one of the smaller caves. 



VESPERTILIONIDAE 

Pipistrellus nanus nanus (Peters) 

Vespertilio nanus Feters, 1852, Eeise nach Mossambique, Siiugeth., p. 63, pi. 
xvi, fig. 2: Inhambane, Mozambique. 

4 cf 1 9 (M. C. Z. 39200, 40808-11) Kibale Forest, U. 19. xii. 38. 
2 d" 2 9 (M. C. Z. 40763-6) Idjwi Id., Lake Ivivu, B. C. 4. iii. 39. 
6 d' 4 9 (M. C. Z. 38995-9, 39049-53) Magrotto Mtn., T. T. 29. 

vi. 39. 
10 d" 6 9 (alcoholic specimens) Magrotto Mtns., T. T. 29. vi. 39. 

Native names. Kihuguhugu^ (Lutoro); helihu (Luamba); nundu 
(Kisambara). 

' The Education Department, however, reserves this for fruit bats, and considers Kahundu 
(Lutoro) the correct name for insectivorous bats. 



166 bulletin: museum of comparative zoology 

Discussion. This widely distributed little bat is frequently found by 
day resting inside the rolled-up central frond of the banana plant. 

Measurements, d". 45. 34. 6. 9. 99 mm., 9 • 45. 42. 6. 9. 104 mm., 
the cf being from Magrotto, the female from Idjwi Id. 

MiNioPTERUS minor Peters 

Miniopterus minor Peters, 1867 (1866), Monatsb. Akad. Wiss. Berlin, p. 885: 
Zanzibar coast. 

3 cf 1 9 (M. C. Z. 38946, 38992-4) Siga Caves, Tanga, T. T. 10. 
vi. 39. 

Discussion. These four specimens agree in their dark blackish-gray 
color above, drabby below, becoming dark smoky gray on the throat 
and chest. Forearms, 39-42 mm. ; skull length, 14 mm. The forearm 
of the type is said to measure only 37 mm. On the whole, however, 
these specimens may be best referred to minor, of which they are 
topotypie. 

Measurements, cf . 51. 40. 7. 11. 137 mm., 9 • 47. 43. 8. 10. 138 mm. 

Habitat. Netted at the entrance to one of the smaller caves. 



MOLOSSIDAE 
Mops angolensis orientis subsp. nov. 
6 d' 4 ? (M. C. Z. 38826-35) Kitaya, T. T. 3. iv. 39. 

Type. Museum of Comparative Zoology No. 38829, an adult male, 
skin and skull, from Kitaya, Rovuma River, southeastern Tanganyika 
Territory. Collected by Arthur Loveridge, April 3, 1939. 

Description. One of the stout-bodied forms with white underside, 
slightly smaller than Mops (Allomops) angolensis osborni of the Great 
Lakes region, to which it is related, and with the upper surface faintly 
tinged with tawny ochraceous instead of being uniform drabby 
brown; skull sHghtly smaller and with less development of the sagittal 
crest. 

Distribution of the fur as in M. a. osborni, the wing above naked, but 
with a narrow line of whitish hairs on the anterior and posterior 
sides of the humerus; hind legs practically naked as well as the pos- 
terior part of the rump and the anal region, where, however, minute 
scattered hairs are visible under a lens. Face, chin and ears blackish, 
with minute sparse hairs; toes with the usual longer stiff hairs as well as 
lateral hairs on outer edge of the first and fifth toes. Entire under 
surface of the body pure white to the edge of the membrane, with 



ALLEN AND LOVERIDGE: AFRICAN MAMMALS 167 

slight individual variation, so that in some specimens the sides of the 
neck are pale drab while in others this tint is more extensive, from the 
axilla back along the side halfway to the groin. Wings dusky brown, 
becoming whitish lateral to the forearm; this pale area varies individu- 
ally but may extend to the tip of the third finger. 

Measurements. The collector's measurements of the type are: head 
and body, 86 mm.; tail, 40; hind foot, 13; ear, 19; spread of wings, 
340. 

The cranial measurements of the skull of the type are: greatest 
length, 21,8 mm.; condylobasal length, 19.2; palatal length, 9.9; 
zygomatic wddth, 13.0; mastoid width, 11.5; width outside molars, 
9.5; upper cheek teeth, 7.7. 

In the adult-male skull the occiput is produced behind and squarely 
truncate, with a transverse angular crest, while the knife-like sagittal 
crest extends forward to the interorbital level. In the upper jaw the 
anterior premolar is minute and crowded into the outer angle between 
the canine and large premolar, which are in contact in their median line. 
In the lower jaw the anterior pair of incisors overlaps the posterior pair, 
and all four are bifid. In males the two lower premolars are of nearly 
equal height, but in females the anterior one is distinctly the shorter. 

Remarks. The series is uniform in the decided tint of russet above 
instead of the dull drab of oshorni, while the much clearer and more 
extensive white area of the lower side and the whitish wings, distin- 
guish it further and altogether probably reflect the somewhat different 
climatic conditions of this part of the coast as compared with the 
Tana River and Lake region. The form. Mops faradjius, is a darker 
representative found in the northeastern Congo forest; probably all 
should be regarded as races of M. angolensis. 

Habitat. These bats were roosting under the galvanized roof of the 
baraza house. I set up the net at 6.30 p.m.; the bats began emerging 
at 6.45 p.m. and by 7 p.m. I had removed ten of this species and one 
of the much smaller C. limbatus. Naturally their stomachs were empty, 
none of the females held embryos. The natives apply the same name 
kiputipidi to them as for other small bats. 

Chaerephon LIMBATUS (Pctcrs) 

Dysopes limhatm Peters, 1852, Reise nach Mossambique, Saugeth., p. 56, pi. 
xiv : Mozambique Island and Sena, Mozambique. 

9 (M. C. Z. 38825) Kitaya, T. T. 3. iv. 39. 

Native name. Kiputiputi (Kimakonde); haundo (Kiyao). 



168 bulletin: museum of comparative zoology 

Discussion. This female agrees closely with Peters's figures and de- 
scription. It is difficult to see how C. hindei of Thomas, from Fort Hall, 
Kenya Colony, is really different. 

Habitat. Taken together with the larger Mops as described above. 

Chaerephon pumilus naivashae HoUister 

Chaerephon pumilus naivashae Hollister, 1916, Smithsonian Misc. Coll., 66, 
No. 1, p. 4: Naivasha Station, Kenya Colony. 

1 (alcoholic) Ibis Hotel, Jinja, U. 4. xi. 38. 
4 d' 5 9 (M. C. Z. 39347-55) Budongo Forest, U. 24-26. xi. 38. 
2 cf 20 9 (alcoholic) Bisu, Budongo Forest, U. 26. xi. 38. 

Discussion. This series is very uniform in its characters. The color 
is drab brown above, paler and more drab below; a narrow line of 
white fur borders the body on the ventral side from axilla to groin and 
is continued across the upper thigh; membranes dusky, more or less 
translucent at the sides. In size the series is intermediate between the 
measurements given by Hollister (1918, p. 98) for this race and the 
typical form from Eritrea and the Sudan, having the small forearm of 
the latter (37-39 mm. as against 39-42 mm. in naivashae), but with the 
larger skull of naivashae (greatest length 16 mm. or slightly more, 
against 15 mm. or less as in the Red Sea form). The small upper an- 
terior premolar is well developed and fully in place in the tooth row. 

Measurements. &. 55. 43. 6. 10. 133 mm., 9 • 60. 37. 7. 16. 140 mm. 

Breeding. On November 26, each of the five females examined, held 
an embryo of which some were well advanced. 

Habitat. Very abundant in the buildings of Buchanan's Saw-Mills. 
At dusk I spread a net between two baulks of timber, and on returning 
twenty minutes later, found six males and twenty-five females en- 
tangled. The entire net was transferred to a cyanide tin, so that later 
it was possible to disentangle the corpses with comparative ease. 

MUSTELIDAE 

Poecilogale albinucha doggetti Thomas & Schwann 

Poecilogale doggetti Thomas & Schwann, 1904, Abstr. Proc. Zool. Soc. London, 
No. 6, p. 22: Eurumba, Ankole, Uganda. 

c^ (M. C. Z. 39357) Mushongero, U. 1. ii. 39. 
c? 9 (M. C. Z. 39140-1) Kisenyi, B. R. 10. ii. 39. 

Native name. Samonyiga (Lugezi). 

Discussion. This large race represents the westward extreme of the 



ALLEN AND LOVERIDGE: AFRICAN MAMMALS 169 

species in Central Africa. Lonnberg has recorded specimens from 
Rutshuru, regarding them as a race of the South African P. alhinucha, 
rather than a distinct species. He mentions the discrepancy in size 
between males and females, but without giving measurements. Of the 
three specimens secured by Loveridge, an adult male from Mushon- 
gero, Lake Mutanda, and an adult female (M. C. Z. 39141) with well- 
developed sagittal crest, from Kisenyi, Lake Kivu, show the following 
skull measurements, respectively : greatest length, 59.5, 53.8 mm.; basal 
length, 55.6, 51.6; palatal length, 27.2, 25.6; zygomatic width, 32.6, 
29.5; mastoid width, 29.6, 26.1; width outside upper large premolars, 
19.8, 19.0; upper tooth row, 19.0, 17.8; lower tooth row, 19.3, 17.9. 
The skull of the female appears much smaller and more slender than 
that of the male, notwithstanding that the actual differences are not 
very great. 

Measurements, cf . 330. 210. 45. 18 mm., 9 • 325. 130. 39. 16 mm., 
from Mushongero and Kisenyi respectively. 

AoNYX CAPENSis ?HiNDEi (Thomas) 

Lutra capensis hindei Thomas, 1905, Ann. Mag. Nat. Hist. (7), 15, p. 78: Fort 
Hall, Kenya Colony. 

1 (M. C. Z. 39426) Bought at Nyakabande, U. 4. ii. 39. 

Discussion. This native-made skin has the head and shoulders 
hoary, the throat sharply contrasted white as far as a line from lower 
rim of eye- to just below the ear. It is provisionally referred to the 
race hindei, the validity of which may be still open to doubt. 

Remarks. Nyakabande, Kigezi district, lies between Lakes Mutanda 
and Bunyonyi. The skin was one of two bought for $3.25 (the other 
being in the ethnological collections of the Peabody Museum) in the 
hopes that it might represent Paraonyx phillipsi Hinton, from Lake 
Bunyonyi. It lacks lips and vibrissae, however, these having been cut 
away; nor does it exhibit the median head stripe of phillipsi. The na- 
tive from whom it was purchased said that it came from the south end 
of Lake Mutanda where this larger otter lives in the vast papyrus 
swamps. All the natives agreed that only two species of otter occur in 
the lake. 

LuTRA MACULICOLLIS TENUIS Pohle 

Lutra tenuis Pohle, 1920 (1919), Arch. Naturg., 85, A, p. 53: Lake Mohasi, 

Belgian Buanda-Urundi. 
Lutra maculicollis mutandae Hinton, 1921, Ann. Mag. Nat. Hist. (9), 7, p. 368: 

Lake Mutanda, British Ruanda, Uganda. 

cf (M. C. Z. 39425) Mushongero, L. Mutanda, U. 31. i. 39. 



170 bulletin: museum of comparative zoology 

Native names. Ngonyi (Lukiga); nzihi (Lulega, for kivuayia). 

Discussion. This handsome skin is dark in color and the throat 
spots are sharply contrasted white. The postorbital processes as usual 
are lacking. This specimen is a topotype of mutandae, which is almost 
certainly a synonym of tenuis, described a year earlier from a lake lying 
forty miles further south. 

Measurements, cf. 630. 335. 173. 16 mm. 

Diet. A frog {Xeno-pus I. hunyoniensis) was in its stomach. These 
otters engage in fishing in the early morning and late afternoon, and 
the one listed here was so engaged when I shot it (from a canoe) in the 
back of the head. 

VIVERRIDAE 

Genetta tigrina stuhlmanni Matschie 

Genetlu stuhbnanni Matschie, 1902, Verb. d. V. Internat. Zool. Congress, 
Berlin, 1901, p. 1142: Bukoba, Lake Victoria, Tanganjoka Territory. 

cT (M. C. Z. 39394) Mabira Forest, U. 9. xi. 38. 

Native name. Kasimha (Luganda). 
Measurements, cf . 490. 430. 80. 40 mm. 

Genetta tigrina suahelica Matschie 

Genetta sriahelica Matschie, 1902, Verb. d. V. Internat. Zool. Congress, Berlin, 
1901, p. 1143: East coast of Tanganyika Territory to Mombasa, Ken3'a 
Colony. 

9 (M. C. Z. 38868) Nchingidi, T. T. 11. v. 39. 

Discussion. This is a youngish animal from the Rondo Plateau, 
southwest of Lindi. 

Myonax sanguineus PROTEUS (Thomas) 

Mungos gracilis protetis Thomas, 1907, Ann. Mag. Nat. Hist. (7), 19, p. 119: 
Mubuku Valley, Euwenzori Mountains, Uganda, 7000 ft. 

o (M. C. Z. 39272) Mihunga, U. 29. xii. 38. 

Native names. Kasisi (Lukonjo); ^kasindi (Lutoro). 
Discussion. This topotype is an immature individual with blackish 
limbs and terminal third of tail, but with the head and back finely 

> Somewhat doubtful as applied also to a sqxiirrel. 



ALLEN AND LOVERIDGE: AFRICAN MAMMALS 171 

sprinkled with ochraceous; the ventral side of the body is dark smoky 
with very Uttle speckling. 

Measurements. 9 . 320. 240. 51. 25 mm. 

Parasites. Larval pentastomids (Armillifer annilhdiis) in mesen- 
tery. 

Atilax paludinosus ?mordax (Thomas) 

Mungos paludinosus mordax Thomas, 1912, Ann. Mag. Nat. Hist. (8), 10, p. 
588: Rombashi River, northwest of north end of Lake Nyasa, Tan- 
ganyika Territory. 

cf juv. (M. C. Z. 39158) Mushongero, U. 31. i. 39. 

Native name. Ckihura (Lukiga). 

Discussion. This specimen from Lake Mutanda is very young with 
only the two central pairs of milk incisors in place. It is melanistic 
with the rather long, lax pelage entirely blackish. The proportionately 
short tail (90 mm. against head and body length of 220 mm.), the 
entirely naked soles, and five-toed feet with unwebbed digits, seem 
sufficient to place it in this genus. 

Measurements, cf juv. 220. 90. 45. 22 mm. 

Bdeogale crassicauda omnivora Heller 

Bdeogale crassicauda omnivora Heller, 1913, Smithsonian Misc. Coll., 61, No. 
13, p. 12: Mazeras, Kenya Colony. 

9 (M. C. Z. 39416) Magrotto Mtn., T. T. 12. vii. 39. 

Native names. Kicheche or 7igogo (Kisambara). 

Discussion. This adult female is melanistic, nearly black all over, 
with the sides of the head and body, and to a less extent the back, 
sprinkled with minute pale tips; the under fur is soiled grayish white, 
the cylindrical tail deep black. The skull agrees closely in measure- 
ments wdth those published by HoUister (1918, p. 135) for the type and 
topotype from Mazeras, not far distant in southeastern Kenya Colony. 

Measurements. 9 . 450. 220. 78. 38 mm. 

Enemies. I came upon a hunting party of four Wasambara and half- 
a-dozen curs, who had located this mongoose at the end of its burrow 
between two rocks in the forest. They had already dug out the burrow 
as far as the rocks would permit and were poking a stick down the 
hole. Suddenly the mongoose burst from its retreat, was seized by a 
dog, then rescued by a man who dealt both animals a blow with his 
heavy stick. The men subsequently ate the mongoose! 



172 bulletin: museum of comparative zoology 

FELIDAE 

Felis brachyura pantasticta Pocock 

Felis servalina pantasticta Pocock, 1907, Proc. Zool. Soc. London, p. 665, pi. 
xxxviii, fig. 3: Entebbe, Uganda. 

1 (M. C. Z. 40840) near Mabira Forest, U. xi. 38. 

Discussion. This native skin, presented to the Expedition by Mr. 
J. L. Jarvis, agrees well in color pattern with those from the north- 
eastern Congo as figured by J. A. Allen (1924, pi. xlvii). It comes from 
a point about sixty miles northeast of the type locality — Entebbe. 

LORISIDAE 

Galago crassicaudatus lasiotis Peters 

Galago lasiotis Peters, 1876, Monatsb. Akad. Wiss. Berlin, p. 912, fig. 1: 
Mombasa, Kenya Colony. 

5 (f 1 9 (M. C. Z. 39408-12, 39415) Siga Caves, Tanga, T. T. 
9-15. vi. 39. 
2 d" (M. C. Z. 39413-4) Amboni Estate, Tanga, T. T. 20 & 26. 
vi. 39. 

Native name. Komba (Kiswahili and Kisambara). 

Discussion. These eight specimens come from localities which, 
according to Schwarz (1931, p. 45), mark the southern limits of the 
range of this coastal race. Most of them show a decided tinge of cinna- 
mon on the back ; three have white tail-tips ; one has the entire terminal 
half of the tail white; while in another the last third is mixed whitish 
and dark. 

Measurements. &. 300. 360. 90. 50 mm., 9 . 280. 360. 90. 48 mm., 
from Amboni and Siga respectively. 

Diet. Apparently acacia gum, in the stomach of one examined. 

Parasites. Nematodes {Subulura sp. 9 ) and encysted larval screw- 
worms (Armillifer armillatus) . 

Habitat. The entire Siga series were shot in one or other of two large 
acacia trees which they ascended soon after dusk. They could be heard 
approaching the trees through the dense underbrush, for they were 
noisily vocal, and would scold at me until the headlight found their 
glowing eyes. Mr. Tanner, of Amboni Estate, told me that one of 
these galagos was electrocuted on the 30,000 volt high-tension wires 
which cross his driveway. The species was heard calling on Magrotto 
Mountain, but I failed to get any there. 



ALLEN AND LOVERIDGE: AFRICAN MAMMALS 173 

Galago demidovii thomasi Elliot 

Plate 3, fig. 1. 

Galago (Hemigalago) thomasi Elliot, 1907, Ann. Mag. Nat. Hist. (7), 20, p. 189: 
Fort Beni, Semliki River, Belgian Congo. 

d" (M. C. Z. 38916) Idjwi Island, B. C. 28. ii. 39. 

Native Jiame. Luhololo (Lulega). 

Discussion. This single specimen, which is referred to the race 
thomasi, is dark brown above, and yellowish-washed below. Galagos 
of this species are especially characteristic of the West African forest 
area and become rarer on its eastward extension, where their place is 
taken by the G. senegalensis and G. crassicaudatus groups, of the 
gallery and savannah forests. 

Measurements, d'. 150. 200. 53. 25 mm. 

Gal.\go senegalensis zanzibaricus Matschie 

Galago zanzibaricus Matschie, 1893, Sitzb. Ges. Naturf. Freunde, Berlin, p. Ill : 
Yambiani {not Muyuni as given), Zanzibar. 

3 cf 2 9 (M. C. Z. 38911-5) Amboni Estate, T. T. 19-24. vi. 39. 

Discussion. According to Schwarz's review (1931, p. 55), specimens 
from the Tanga district are indistinguishable from those of Zanzibar 
and the coast region as far south as Dar es Salaam, but inland in the 
drier country are represented by the race braccatus. 

Measureme7its. c^. 170. 215. 60. 35 mm., 9 . 160. 225. 60. 37 mm. 

Habitat. Exceedingly abundant in the preserved patch of secondary 
forest in the middle of the estate. They were joined by a young G. s. 
moholi which I had brought from Nchingidi a month before, and which 
escaped from my tent at the forest-edge. 

Galago senegalensis moholi A. Smith 

Galago moholi A. Smith, 1836, Rep. Exped. Explor. Cent. Africa, p. 42: Banks 
of the Marikwa and Limpopo, Bechuanaland. 

1 c^ 2 9 (M. C. Z. 38875-7) Nchingidi, T. T. 20. v. 39. 

Discussion. Although in coloration of the limbs and body closely 
similar to the series of zanzibaricus, these skins from the Rondo 
Plateau forest have much darker tails with the terminal third blackish, 
instead of at most the terminal third dusky. Schwarz (1931, p. 57), in 
his review, includes as of this race specimens from Liwale, Kilwa dis- 



174 bulletin: museum of comparative zoology 

trict. One would expect a certain amount of intergradation in this 
general area. 
Breeding. A young one accompanied one of the females. 



CERCOPITHECIDAE 

Cercocebus albigena johnstoni (Lydekker) 

Semnocebus albigerui johnstoni Lydekker, 1900, Novit. Zool., 7, p. 594: "Near 
Lake Tanganyika," but probably from the Semliki or Ituri Forest (fide 
Lorenz). 

cf (M. C. Z. 39402) Mabira Forest, U. 12. xi. 38. 
3 d' (M. C. Z. 39388, 39395-6) Kibale Forest, U. 9-15. .xii. 38. 

Native name. Sewagaba (Luganda). 

Measurevients. cf. 630. 840. 175. 40 mm., from Kibale Forest. 

Diet. Stomach full of finely masticated green matter. 

Habits. The Mabira male was in the company of a troupe of young 
C. n. schmidti. In Kibale forest, on the other hand, where these manga- 
beys were numerous as well as tame, they went about in large bands of 
their own kind. 

Cercopithecus nictitans SCHMIDT! Matschie 

Cercopithecus schmidti Matschie, 1892, Zool. Anz., 15, p. 161 : Manyema, west 
of north end of Lake Tanganyika, Belgian Congo. 

9 (M. C. Z. 39380) Mabira Forest, U. 11. xi. 38. 
cf (M. C. Z. 39374) Budongo Forest, U. 24. xi. 38. 
cf 9 (M. C. Z. 39379, 39385) Kibale Forest, U. 10 & 13. xii. 38. 

Discussion. All four specimens agree closely in color pattern. 
Those from the Kibale Forest are practically topotypes of mpangae 
Matschie (for the Mpanga River flows through the Kibale Forest) and 
have been compared with the topotypic series of kaimosae Heller from 
which they do not appear to differ. C. schmidti mpangae Matschie, 
together with its synonyms, is, therefore, referred to the synonymy of 
C. n. schmidti with a distribution from the eastern Belgian Congo 
across Uganda to western Kenya Colony. 

Measurements. &. 530. 870. 145. 35 mm., 9 . 495. 615. 120. 30 mm., 
from Budongo and Mabira Forests respectively. There is considerable 
variation in tail length. 

Diet. A pea-like fruit was present in the stomach of the Budongo 
male. 



ALLEN AND LOVERIDGE: AFRICAN MAMMALS 175 

Habits. This Budongo male was feeding in the same tree with a 
solitary female of C. m. stuhlmanni. When disturbed, these white- 
nosed, red-tailed monkeys maintain a continuous bird-like chirping. 

Cercopithecus l'hoesti l'hoesti Sclater 

Cercopithecus l'hoesti P. L. Sclater, 1899 (1898), Proc. Zool. Soc. London, p. 586, 
pi. xlviii: Congo. 

2 9 (M. C. Z. 39382, 39386) Ivibale Forest, U. 10 & 13. xii. 38. 

Discussion. This handsome, red-backed, black-limbed monkey with 
contrasting white throat and side whiskers, probably does not extend 
much to the eastward of this region. 

Measurevients. Larger 9 . 600. 650. 160. 30 mm. The relatively 
short tail appears to be a good distinguishing feature. 

Parasites. Hair-like nematodes ( Trichuris trichuria) were present in 
the stomach. 

Cercopithecus mitis stuhlmanni Matschie 

Cercopithectis stuhlmanni Matschie, 1893, Sitzb. Ges. Naturf. Freunde Berlin, 
p. 225: North of Ivinyawanga, northwest of Lake Albert, Belgian Congo. 

3 9 (M. C. Z. 39375, 39377-8) Budongo Forest, U. 24. xi-2. xii. 38. 
2 cf (M. C. Z. 39398, 39401) Mubuku Valley, U. 2-4. i. 39. 

1 c^ (M. C.'Z. 39399) Mihunga, RuwenzorLMtns., U. 13. ii. 39. 

Native names. Nkima (Lukonjo and Lutoro). 

Discussion. This is a common species in western Uganda. The 
three males from Ruwenzori at 7000 feet are practically topotypes of 
the supposed race carruthersi, described from 10,000 feet, now regarded 
as a synonym. 

Measurements, cf . 630. 800. 160. 40 mm., 9 . 510. 785. 138. 34 mm., 
from Mubuku and Budongo respectively. 

Diet. The stomachs of the Mubuku males were full of berries. 

Habits. Both the fine Mubuku males were quite solitary, and the 
only examples of their species seen during our stay in the valley. One 
of the Budongo females, however, was associated with a male C. n. 
schmidti. 

Cercopithecus mitis schoutedeni Schwarz 

Cercopithecus leucampyx schoutedeni Schwarz, 1928, Revue Zool. Bot. Afri- 
caine, 16, p. 126: Idjwi Island, Lake Kivu, Belgian Congo. 

2 d" (M. C. Z. 38376, 39387) Idjwi Island, B. C. 20-21. ii. 39. 
Native name. Nchivia (Lulega). 



176 bulletin: museum of comparative zoology 

Discussion. These two topotypes amply bear out the characteristics 
of this race, which has a much paler, slightly buflFy back as compared 
with the darker race stuhlmanni farther northward. A similar differ- 
ence is noticeable in a number of other mammals, in which the races 
about Lake Kivu are paler than their representatives in the Ru- 
wenzori region. 

Measurements. Larger cf . 600. 860. 172. 36 mm. 

Diet. Stomach of one full of pumpkin. In addition to pumpkin, I 
observed that these monkeys eat maize, millet, and mahoga leaves by 
raiding gardens adjacent to the forest. 

Enemies. If the natives did not encroach upon the forest as the^'^ 
are doing, or cultivate in its immediate vicinity, there might be less 
plundering. As things are there is much antagonism between them, 
and the monkeys are very wild through much hunting. On several 
occasions I heard parties of Bambuti hunting them in the forest to the 
accompaniment of an infernal din. As, with the exception of the serval, 
these monkeys are the only large mammal on the island, the natives are 
eager to eat them. 

Cercopithecus mitis monoides Geoflfroy 

Cercojyithecus monoides I. Geoffrey, 1841, Arch. Miis. d'Hist. Nat. Paris (1), 2, 
p. 558, pi. xxxi: Lectotype, Rufigi River, 8° S., Tanganyika Territory. 

2 9 (M. C. Z. 39383-4) Mikindani, T. T. 11 & 14. iv. 39. 
9 (M. C. Z. 39390) Siga Caves, Tanga, T. T. 9. vi. 39. 
9 (M. C. Z. 39389) Magrotto Mtn., T. T. 10. vii. 39. 

Native names. Unima (Kimakonde); lichima (Kimwera); ndoue 
(Kisambara). 

Discussion. This is a somewhat pale coastal form, with pale-orange 
back, gray hind limbs and under surface; the forearm and thumb are 
deep black, the hind feet black and gray speckled. 

Measuremenis. 9 . 480. 670. 125. 30 mm. From Magrotto. 

Diet. On Amboni Estate I was shown young sisal shoots destroyed 
by these monkeys. In consequence a bounty of one shilling is paid for 
each monkey killed. 

Remarks. Seen also at Kitaya; and heard at Mbanja, Nchingidi and 
Amboni where they are shy through much hunting. 

Cercopithecus aethiops centralis Neumann 

Cercopithccnti centralis Neumann, 1900, Zool. Jahrb., Syst., 13, p. 533: Bukoba, 
Lake Victoria, Tangan>aka Territory. 

1 cf 2 9 (M. C. Z. 39381, 39403, 39418) Mihunga, U. 10 & 17. i. ^9. 



ALLEN AND LOVERIDGE: AFRICAN MAMMALS 177 

Native name. Nkendc (Lukonjo and Lutoro). 
Measuremenis. d'. 500. 520. 132. 35 mm., 9 • 500. 550. 125. 38 mm. 
Breeding. The female was carrying a young 9 • 225. 360. 85. 31 mm., 
on January 17. 

Enemies. The bodies were requested by the Bakonjo, who eat them. 

Cercopithecus aethiops johnstoni Pocock 

Cercopithecus pygerythrus johnstoni Focock, 1S07, Froc. Zool. Soc. London, p. 
638: Moshi, near Kilimanjaro, Tanganyika Territory. 

9 (M. C. Z. 39393) Amboni Estate, near Tanga, T. T. 24. vi. 39. 

Native name. Tumbili (KiswahiH and Kisambara), 
Measurements. 9 • 450. 600. 120. 32 mm. 

COLOBIDAE 

CoLOBUs POLYKOMOS PALLiATUs Peters 

Colobus palliatus Peters, 1868, Monatsb. Akad. Wiss. Berlin, p. 637: East 
Africa, opposite Zanzibar, i.e. Fangani Eiver, Tanganyika Territory. 

cT 9 (M.C.Z. 39391-2) Amboni Estate, near Tanga, T.T. 22. vi. 39- 

Native name. Mhega (Kiswahili and Kisambara). 
Discussion. This is the form of the coastal ranges of northern 
Tanganyika and southern Kenya Colony. 

Measurements, cf . 570. 730. 167. 31 mm., 9 • 500. 570. 140. 29 mm. 

Colobus badius tephrosceles Elliot 

Colobus tephrosceles Elliot, 1907, Ann. Mag. Nat. Hist. (7), 20, p. 195: Ruahara 
River, Toro, east side Ruwenzori Mountains, Uganda, 4000 ft. 

2 d" (M. C. Z. 39397, 39400) Kibale Forest, U. 10. xii. 38. 

Native name. Kobi (Luganda). 

Discussion. These examples of tephrosceles from Toro must be al- 
most topotypic; although generally rare in collections, Loveridge 
found it abundant in the Kibale Forest. This race is nearest to C. b. 
rufomitratus, the range of which is now confined to the gallery forests 
of the lower Tana River. 

Measurements, cf. 600. 720. 160. 31 mm. 

Habits. The troupes of red-capped colobus were very large, at least 
fifty, perhaps double that number, of animals forming a troupe. When 
fired at they uttered very threatening cries. 



178 bulletin: museum of comparative zoology 

MANIDAE 

Phataginus tricuspis mabirae subsp. nov. 

Plate 3, fig. 2, and Plate 4, figs. 1-2. 

Type. Museum of Comparative Zoology, No. 39417, an adult male, 
skin and skull, from Mubango, Mabira Forest, Chagwe, Uganda, 
collected by Arthur Loveridge, November 12, 1938. 

Description. In typical P. tricuspis from the Cameroons, the nasals 
usually show a constriction or waist at the middle then abruptly ex- 
pand posteriorly and are widest at the point of junction between the 
maxillae and the anterior corner of the frontals, where the nasals show 
a sharp angle; in the Uganda skull, however, they are not so narrowed, 
but expand evenly and gradually with very little trace of an angle, 
and are produced farther back behind the maxillo-frontal contact. 
In all four Cameroons skulls examined, the transverse suture formed 
by the anterior outline of the parietals is farther forward and reaches 
laterally the angle formed by the anterior root of the squamosal pro- 
cess, whereas in the Uganda skull this line falls well posterior to the 
squamosal angle (by about 5 mm.). In the latter too, the premaxilla 
is considerably wider and its ascending process is shorter and much 
less tapering. In dorsal view there is no interorbital narrowing such 
as is obvious in the West African skulls, but the outlines are nearly 
straight and diverging from the anterior tip. 

The thin hairy coat on the underside of the body is different in color 
from any of the eight other specimens from western Africa seen, an 
'orange cinnamon' instead of dark blackish brown or gray. The tips 
of the scales are truncate through wear. 

Measurements. The collector's measurements are: head and body, 
355 mm.; tail, 555; hind foot, 47. The skull measures: greatest length 
(which is same as condylo-incisive length), 79 mm.; palatal length, 
43.7; mastoid width, 34.3; lacrimal width, 22.0; width at fronto- 
parietal suture, 29.4; length of mandible, 54.5. 

Remarks. In studying the series of Phataginus collected by the 
American Museum's Congo Expedition, Dr. Robert T. Hatt (1934, 
p. 655) called attention to the wide individual variation in the confor- 
mation of the bones of the skull and in other characters, so that the 
significance of the details of difference mentioned above may to some 
extent be less than supposed; nevertheless they seem sufficiently 
striking on the whole to justify the distinction of the extreme eastern 



ALLEN AND LOVERIDGE : AFRICAN MAMMALS 179 

animal of Uganda from that of West Africa, as represented by Camer- 
oons and Ivory Coast specimens. 

Native name. Lugave (Luganda). 

Parasites. Ticks {Amblyomma cuneatum) were numerous. 

SCIURIDAE 

Aethosciurus ruwenzorii ruwenzorii (Schwann) 

Sciurus rufobrachiatus ruwenzorii Schwann, 1904, Ann. Mag. Nat. Hist. (7), 
13, p. 71: Wimi Valley, Ruwenzori Mountains, Uganda, 7880 ft. 

1 d^ 2 9 (M. C. Z. 39277-9) Mihunga, U. 13-16. i. 39. 

Native name. Kasindi (Lukonjo and Lutoro). 

Discussion. These greenish-oHve squirrels with white ventral stripe 
from chin to anus, from eastern Ruwenzori, are almost topotypic. 

Measurements, c^. 200. 230. 51. 18 mm., 9 . 225. 230. 51. 18 mm. 

Breeding. A note to the effect that they were 'not breeding' was 

made by the collector, but the senior author considers them to be in 

. nursing condition with six mammae — two pectoral and four inguinal. 

Heliosciurus rufobrachium arrhenii Lonnberg 

Heliosciurus rufobrachiatus arrhenii Lonnberg, Kungl. Svenska Vet.-Akad. 
Handl., Stockholm (2), 58, No. 2, p. 68: Masisi, near Kivu, Belgian Congo. 

4 o^ 4 9 (M. C. Z. 39097-102, 39138-9) Idjwi Id., B. C. 17-28.11. 39. 

Native name. Lisbeshi (Lulega). 

Discussion. This series of skins from Lake Kivu shows a slight inten- 
sification of the rusty rufous on fore and hind limbs in comparison 
with those from Uganda referred to the race nyansae. The backs are 
similar in the two, but the long hairs of the tail are mainly black with 
longer white tips and two or three narrow white, instead of buffy, 
rings. One individual, however, shows the middle ring ochraceous 
distally. They are assumed to represent arrhenii which probably 
intergrades with nyansae and other neighboring races. 

Measurements. &. 252. 240. 52. 19 mm., 9 . 245. 190. 55. 20 mm. 

Heliosciurus rufobrachium nyansae (Neumann) 

Sciurus nyansae Neumann, 1902, Sitzb. Ges. Naturf. Freunde Berlin, p. 56: 
Kwa Kitoto, Kavirondo, Kenya Colony. 

cf (M. C. Z. 39276) Mabira Forest, U. 15. xi. 38. 

2 9 (M. C. Z. 39274-5) Budongo Forest, U. 19 & 25. xi. 38. 



180 bulletin: museum of comparative zoology 

Native name. Kakerebive (Luganda). 

Measurements. &. 220. 225. 53. 20 mm., 9 . 220. 250. 50. 16 mm. 

Heliosciurus undulatus undulatus (True) 

Sciurus undulatus True, 1892, Froc. U. S. Nat. Mus., 15, p. 465, fig. 3: Mt. 
Kilimanjaro and Kahe, south of it, Tanganjaka Territory. 

1 d" 1 9 (M. C. Z. 39107-8) Siga Caves, near Tanga, T. T. 9. vi. 39. 
6 cf 9 9 (M. C. Z. 39103-6,-9-19) Magrotto Mtn., T. T. 30. vi-15. 
vii. 39. 

Native name. NIcenda (Kisambara). 

Discussion. The series from Magrotto is referred to the typical 
race. There is much individual variation in the amount of rusty on the 
under surface. In a few individuals this region is nearly uniform drabby 
brown, in others the inner side of the tibiae is bright rufous, less so on 
the under side of the forearms, while others again show intermediate 
conditions. 

The two from Tanga district agree in being a trifle paler below than 
the Magrotto series, their throats lacking a distinct rufous tinge, being 
a pinkish drab instead; while on the sides of the back the pale sub- 
terminal ring of the separate hairs is white, instead of ochraceous buff, 
giving the flanks and dorsum a pale hoary appearance. They do not, 
however, conform to Thomas's description of his race daucinus from 
Mombasa, as one might have supposed. 

Measurements, d'. 250. 280. 54. 20 mm., 9 • 237. 300. 50. 22 mm. 
From Magrotto and Siga Caves respectively. 

Diet. The Siga pair were feeding in a wild fig. Most of the Magrotto 
series were shot between 8 and 9 a.m. while eating nuts of the oil palm; 
one fell dead with a nut in its jaws! Doubtless their abundance on the 
Estate may be attributed to the rich food supply furnished by the 
palm plantation. Stomach contents of one consisted solely of these 
yellow nuts, that of another a white cheesy substance, a third's was 
mostly green matter. 

Parasites. None observed ! 

Enemies. Eaten by the Wasambara. 

EUNISCIURUS PYRRHOPUS VICTORIAE Subsp. nOV. 

Type. Museum of Comparative Zoology, No. 39199, an adult male, 
skin and skull, from Kibale Forest, Toro, Uganda, collected by Arthur 
Loveridge, December 16, 1938. 



ALLEN AND LOVERIDGE: AFRICAN MAMMALS 181 

Description. Nearest to F. p. akka of northeastern Belgian Congo, 
its closest neighbor, but differing in having the entire lower surfaces of 
the body and upper parts of limbs strongly suffused with ochraceous, 
the hairs at their bases white. In F. p. akka the entire under surface is 
white. 

Dorsally, the central region of the body from between the eyes to 
the base of the tail is a mixture of all-black hairs with those having a 
narrow subterminal ring of ochraceous buff and a minute black tip; 
below this area on each side is a narrow stripe of clear ochraceous buff 
running from shoulder to hip; below this again on the flanks the bases 
of the hairs become slaty gray with much broader ochraceous rings 
resulting in a lateral band of clearer ochraceous buff and gray. The 
upper lips, a spot at the base of the vibrissae extending to the eye ring, 
clear orange rufous continuing broadly behind the eye to the base of 
the ear, but darker and duller rufous. Scrotum gray. Tail above, 
black, many of the hairs white-tipped forming a narrow fringe; under 
side with the central area dull rufous, bordered by black and narrowly 
fringed with white. Ears short, with their upper rims and posterior 
lobe black, edged ventrally with pale ochraceous. 

Measurements. The collector's measurements are: head and body, 
225 mm.; tail, 160; hind foot, 46; ear, 17. The skull measures: greatest 
length, 49.0 mm.; basal length, 40.7; palatal length, 24.6; zygomatic 
width, 25.2; mastoid width, 19.8; width across molars, 11.0; upper 
cheek teeth, 8.5; lower cheek teeth, 8.5: 

Remarks. This is the most eastern of the described races of F. 
pyrrhopns, and though unfortunately based on but a single specimen, 
nevertheless differs so strikingly in its bright ochraceous under side 
from its nearest neighbor, F. p. akka from Monbuttu, that it seems 
worthy of a name. According to J. A. Allen's (1922, p. 54) account, 
akka is "exceedingly constant in coloration," his entire series of thirty- 
two skins having the under parts "pure white to the base of the fur." 

Habitat. This little squirrel ran across the new road just cut through 
the forest southeast of Fort Portal, during a heavy downpour, then 
paused among the brushwood piled up at the side. 

Paraxerus palliatus frerei (Gray) 

Macroxus annulatus var. frerei Gray, 1873, Ann. Mag. Nat. Hist. (4), 12, p. 
265: Zanzibar. 

cf (M. C. Z. 39096) Siga Caves, T. T. 12. vi. 39. 
Discussion. This squirrel, shot at the entrance of the caves near 



182 bulletin: museum of comparative zoology 

Tanga, on the mainland almost opposite the type locality, closely 
matches topotypical material. The name suahelicus, formerly used for 
the mainland animal, is now regarded as a synonym. 
Measurcvicnts. cf . 195. 235. 52. 20 mm. 

Paraxerus sponsus bridgemani Dollman 

Paraxerus bridgemani Dollman, 1914, Ann. Mag. Nat. Hist. (8), 14, p. 152: 
Indook, Panda, Mozambique. 

2 cf 2 ? (M. C. Z. 38883, -5-6, -90) Kitaya, T. T. 28. iii-4. iv. 39. 
2 cf 2 9 (M. C. Z. 38884, 38887-9) Mikindani, T. T. 10 & 20. iv. 39. 
1 cT 1 9 (M. C. Z. 38881-2) Nchingidi, Rondo, T. T. 9-12. v. 39. 

Native names. Kivuki (Kiyao); kikuhi (Kimakonde at Kitaya); 
chikuvi (Kimakonde at Mbanja); chiruma (Kimwera). 

DiscussioJi. In general the series agrees well with the original de- 
scription, but for a certain amount of individual variation in the 
amount of orange red in the tail. This may form a lateral border as 
originally described, or may occur as a mixture with the blackish hairs 
on both upper and lower sides. In some, with wear, it has disappeared 
altogether. 

Measurements, cf . 200. 185. 42. 21 mm., 9 • 200. 115. 45. 18 mm., 
from Kitaya and Nchingidi respectively. 

Breeding. On April 3 and 20, half to two-thirds grown young were 
shot, the smallest, a cf , measuring 120. 125. 36. 15 mm. 

Habitat. At Kitaya two were shot on paths, two in the trees, a 
Mikindani squirrel in a mango tree. The species is addicted to chatter- 
ing in the undergrowth very persistently; it is unusually wary, however, 
about exposing itself. 

Paraxerus ochraceus aruscensis (Pagenstecher) 

Sciurus cepapi var. aruscensis Pagenstecher, 1885, Jahrb. Hamburg. Wiss. 
Anst., 2, p. 42: Great Arusha and Pangani River, Tanganyika Territor}\ 

cf 9 (M. C. Z. 38908, 38910) Siga Caves, T. T. 9 & 12. vl. 39. 
9 (M. C. Z. 38909) Amboni Estate, Tanga, T. T. 19. vi. 39. 

Discussion. The three specimens agree in having the belly broadly 
buffy, in contrast to typical ochraceus of Bagamoyo in which the 
mixed gray of the sides encroaches on the middle part of the belly. 

Measurements, d". 175. 130. 39. 18 mm., 9 • 168. 185. 38. 16 mm., 
from Siga and i\.mboni respectively. 

Habitat. Both Siga squirrels were shot in acacia trees, the Amboni 
animal was driven out of piled-up rubbish by a tractor. 



ALLEN AND LOVERIDGE: AFRICAN MAMMALS 183 



Paraxerus flavivittis exgeanus Hinton 

Paraxerus flavimttis exgeanus Hinton, 1920, Ann. Mag. Nat. Hist. (9j, 5, p. 311 : 
Kilwa Kisiwani, Tanganyika Territory. 

8 cf 2 9 (M. C. Z. 38891-900) Kitaya, T. T. 27. iii-3. iv. 39. 
1 d" juv. (M. C. Z. 38901) Mbanja, near Lindi, T. T. 1. v. 39. 

3 cf 2 9 (M. C. Z. 38902-6) Nchingidi, T. T. 12-19. v. 39. 

1 c^ (M. C. Z. 38907) Lindi, T. T. 2. vi. 39. 

Native names. Lileje (Kiyao); uhindi (Kimakonde). 

Measurements, d". 193. 162. 41. 20 mm., 9 • 185. 170. 38. 18 ram., 
from Kitaya and Nchingidi respectively. 

Breeding. At Mikindani, on April 1, a nest, composed of a large 
loose assemblage of coconut fibre with some admixture of grass, holding 
a single young one the size of an adult Mns musculus, was found in a 
hollow tree. On being removed from the nest, the young squirrel set 
up a piercing squeak, so was replaced in the nest in a fork of the tree, 
while, from afar, we watched to see if the mother would respond to its 
cries. Presently she did return, but, finding herself observed, fled in- 
continently down the tree and across a field of short grass towards a 
big tree fully a hundred yards away. Surprised that fear should dom- 
inate maternal affections to such an extent,- we returned the nest 
and its young one to its original hole and left them. 

Habits. These squirrels may be seen best an hour after sunrise when 
they bask in close proximity to their holes, into which they disappear 
the moment that they realize that they have attracted attention. 
Their abundance at Kitaya may be attributed to the numerous hard- 
wood trees, full of cavities, which are scattered among the native plots 
of millet. Thus the squirrel, when plundering the millet, is near a 
refuge from which it cannot be easily dislodged, the hardness of the 
wood defying the native axes. The disproportion of the sexes col- 
lected at Kitaya, suggests that the females are either more wary than 
the males, or that litters at this season were monopolizing their atten- 
tion. 

Tamiscus alexandri (Thomas & Wroughton) 

Funisciurus alexandri Thomas & Wroughton, 1907, Ann. Mag. Nat. Hist. (7), 
19, p. 376: Gudima, Iri River, Upper Uele, Belgian Congo. 

1 9 (M. C. Z. 40796) Mabira Forest, U. 11. xi. 38. 

4 d" (M. C. Z. 39343-6) Budongo Forest, U. 6. xii. 38. 

Native name. Kakerebwe (Luganda). 



184 bulletin: museum of comparative zoology 

Discussion. The white-rimmed ears at once distinguish this striped 
squirrel. In a series of nineteen specimens from the upper Congo, 
J. A. Allen (1922, p. 58) found that in "November, December, January, 
and February specimens the black and white stripes are sharply de- 
fined but in April, May and October they are usually much less dis- 
tinct owing to fading and wear." In the November-December series 
secured by Loveridge, the black stripe on either side of the dorsal area 
is clear and distinct with the trace of a shorter dark stripe just below 
the lateral white line, except in one of the Budongo skins in which the 
black stripes are much dulled by ochraceous-tipped hairs. Two other 
skins from near Beni, northeastern Congo, taken December 23 and 28, 
respectively, have these stripes similarly obscured while a third very 
young one (taken December 13) has both inner and outer black stripes 
sharply defined. It seems more likely that instead of being due to 
"fading and wear," this difi^erence means rather that there are two dis- 
tinct pelages, roughly corresponding to a summer and a winter pelage, 
the duller one representing the winter pelage of more northern lati- 
tudes. A similar dulling of the stripes is seen in the winter examples of 
the striped squirrels, Tamiops, of Asia. In the first coat of the young 
these stripes are contrastingly black. That there seems to be consider- 
able irregularity in the season of assumption of the one or the other, 
probably is due to the fact that the breeding season extends over a 
considerable period, causing a corresponding lack of uniformity in the 
time of change from one coat to the other. 

Measurements, d^. 112. 116. 27. 13 mm., 9 . 112. 125. 25. 12 mm. 

Habitat. The Budongo series were all shot about 9 a.m. in a large 
tree in the forestry nursery at Bisu. Rat traps, baited with bread, 
which had been set about the base of the tree for days previously, 
failed to attract them. 



Tamiscus emini emini (Stuhlmann) 

Sciurus emini Stuhlmann, 1894, Mit Emin Pascha ins Herz von Afrika, p. 320: 
Atyangara, Semliki Eiver, Uganda. 

2 9 (M. C. Z. 40794-5) Mabira Forest, U. 9 & 12. xi. 38. 
2 9 (M. C. Z. 40838-9) Kibale Forest, U. 13 & 16. xii. 38. 

Native name. Kakerebwe (Luganda). 

Discussion. In the series at hand, including the above specimens, 
there is one from Ruwenzori Mountains, a topotype of T. vulcanorum 
lunaris, in a dull pelage with the first black stripe narrower than the 



ALLEN AND LOVERIDGE: AFRICAN MAMMALS 185 

median buffy area; it is dated January. In the others of the series the 
black stripes are sharply contrasted and the first, or subdorsal, stripes 
are broader than the area between them. It seems evident that the 
dull pelage in this and T. alexandri correspond to, and may represent, 
a 'winter' coat whicli has now become slightly irregular in the time of 
its assumption. Possibly also the T. v. lunaris is merely this alterna- 
tive state of T. emini. 

Breeding. On December 13, a 9 held a medium-sized embryo (pre- 
served). 



Protoxerus stangeri centricola (Thomas) 

Sciurus stangeri ce?itricola Thomas, 1906, Ann. Mag. Nat. Hist. (7), 18, pp. 
295, 297: Entebbe, Uganda. 

9 (M. C. Z. 39287) Mabira Forest, U. 18. xi. 38. 
1 d' 2 9 (M. C. Z. 39273, -88, 40791) Budongo Forest, U. 2-3. 
xii. 38. 
9 (M. C. Z. 39286) Kibale Forest, U. 12. xii. 38. 

Native name. Kakerebwe (Luganda). 

Discussion. In all but the Kibale specimen the fingers of the hand 
are uniform black without sprinkling of ochraceous or rufous. The 
throats are darker grayish than in specimens to the westward. 

Measurements, cf juv. 200. 220. 52. 17 mm., 9 . 290. 310. 60. 22 mm., 
both from Budongo. 

Breeding. On December 2, a female was disturbed in the forestry 
nursery as it ran down a tree. It immediately turned about, fled to the 
topmost branches, then lay along a branch where it was fairly exposed, 
but out of gunshot, so I did not fire. As hammering on the trunk did 
not disturb her, a native was posted to keep watch; half-an-hour later 
he reported that she had made for the next tree, thence to a third 
which was still taller. She was restless, however, and kept running to 
and fro from tree to tree until at last she came within range and was 
shot. She held two minute embryos (preserved). The following morn- 
ing as we passed these trees we were scolded by a young male whose 
head protruded from a knot-hole at a height of sixty feet. Striking a 
tree with an axe did not silence it, so again a native was posted to 
summon me when it emerged. When called I found it still scolding and 
jerking its tail; it turned out to be larger than anticipated, viz. cf juv. 
200. 220. 52. 17 mm. 



186 bulletin: museum of comparative zoology 

MUSCARDINIDAE 

Claviglis murinus soleatus (Thomas & Wroughton) 

Plate 5, fig. 3. 

Graphiurus soleatus Thomas & Wroughton, 1910, Trans. Zool. Soc. London, 
19, p. 499: Mubuku Valley, Ruwenzori Mountains, Uganda, 5000-6000 ft. 

a" 9 (M. C. Z. 39267-8) Idjwi Id., B. C. 2. iii. 39. 

Native name. Luleka (Lulega). 

Discussion. This adult male and immature female agree closely with 
the original description. Compared with C. m. saturatus of Mt. Elgon, 
the adult is a darker gray above with only a faint brownish wash, and 
the under surface is smoke gray as described by Thomas and Wrough- 
ton, instead of with whitish-tipped hairs. The immature animal is a 
uniform dark smoke gray above, slightly paler below, with sharply 
contrasted white toes. The braincase appears rather flattened instead 
of vaulted as described for C. vulcanicus of Mt. Karisimbi, the de- 
scription of which otherwise indicates little difference. 

Measurements. &. 123. 92. 17. 14 mm., 9 . 70. 50. 11. 8 mm. 

CRICETIDAE 

Tatera cosensi (Kershaw) 

Taterona cosensi Kershaw, 1921, Ann. Mag. Nat. Hist. (9), 8, p. 567: Vihingo, 
near Ruvu Station, Tanganyika Territory. 

1 cf 2 9 (M. C. Z. 38870-2) Kitaya, T. T. 27. iii. 39. 

3 cf (M. C. Z. 38843-4, -69) Mbanja, T. T. 27. iv. 39. 

Native names. Lipuku (Kiyao); nutu (Kimakonde at Kitaya); nkule 
(Kimakonde at Mbanja, not even generic). 

Discussion. The four adults seem to be referable to this animal, 
which was described from about forty miles inland from Dar es Salaam. 
It closely resembles T. vidua, of which it should probably stand as a 
race, somewhat more mixed with blackish on the middle of the back, 
slightly duller on the sides, and a very little larger in measurements. 

Measurements. &. 175. 152. 35. 20 mm., 9 . 157. 165. 31. 21 mm., 
from Mbanja and Kitaya, respectively. 

Breeding. On April 27, at Mbanja, two young males measured circa 
95. 81. 29. 16 mm. 

Diet. Adults trapped with bread bait. 



ALLEN AND LOVERIDGE: AFRICAN MAMMALS 187 

Tatera nigrita nigrita Wroughton 

Tatera nigrita Wroughton, 1906, Ann. Mag. Nat. Hist. (7), 17, p. 491 : Masindi, 
Unyoro, Uganda. 

cf 9 (M. C. Z. 39216-7) Budongo Forest, U. 25. xi. 38. 

Discussion. These two gerbils, taken at Bisu about twenty miles 
west of the type locaHty, are immature with the last molars not fully 
erupted. In their blackish backs and ears they contrast sharply in 
color with the more buffy T. ruivenzorii. The claws, though actually 
pale horn color, appear black with the caked earth in which they must 
have burrowed. 

Tatera ruwenzorii Thomas & Wroughton 

Tatera ruwenzorii Thomas & Wroughton, 1910, Trans. Zool. Soc. London, 19, 
p. 500: Mokia, s.e. of Ruvvenzori Mountains, Uganda, 3400 ft. 

cT (M.C.Z. 40793) Mabira Forest, U. 11. xi. 38. 

Native name. Fukuzi (Luganda), usually applied to mole rats! 

Discussion. The single specimen agrees closely with the original 
description in its broad interorbital region, long posterior palatal 
foramina, color and measurements. Wroughton found it "very numer- 
ous on the plains around the south end of Ruwenzori," while this ex- 
ample extends the recorded range "slightly to the eastward. 

Measuremeyits. &. 182. 169. 35. 21 mm. 



RHIZOMYIDAE 

Tachyoryctes ruandae Lonnberg & Gyldenstolpe 

Tachyoryctes ruandae Lonnberg & Gyldenstolpe, 1925, Arkiv Zool., 17B, No. 5, 
p. 6: Mt. Muhavura, British Ruanda, Uganda. 

4 c? 9 9 (M. C. Z. 39159, -61-72) Nyakabande, U. 27. i. 39. 
1 ? 4 c^ 7 9 (M. C. Z. 39160, -73-81, -95-6) Mushongero, U. 31. i. 
39. 
1 cf 2 9 (M. C. Z. 39152-4) Kisenyi, B. R. 9-11. ii. 39. 

Native name. Fukuzi (Lukiga). 

Discussion. The first two localities listed above lie a few miles to the 
north of the type locality. Though no mole rats were taken by 
Loveridge on Idjwi Island in Lake Kivu, the Museum has recently 
received a specimen from Nyangesi, 27 km. south of Costermansville, 
and a second from Kanyamundo, which must be very near to the 



188 bulletin: museum of comparative zoology 

southwestern limits of the range of this genus. These two rats were 
collected by Messrs. D. M. Hodgson and W. F. Coultas. 

Adults are russet above, the head nearly black, feet dusky, under sur- 
face dark slaty gray. The young are uniformly slaty gray with scat- 
tered pale-tipped hairs over the posterior part of the back; the russet 
fur first develops along the flanks and spreads upward toward the 
middle of the back. Seven of the thirty specimens have a small white 
area in the middle of the abdomen or lower chest. In ruandae the 
temporal ridges of adults unite in the sagittal line, thus differing from 
ankoliae in which there is said to be always a space of 2-3 mm. between 
them; however, in an adult female from Kisenyi, on the northeast 
shore of Lake Kivu, the ridges are well separated, though united in a 
second specimen from the same locality. The supposed greater size of 
ruandae, as compared with ankoliae of southern Uganda, probably 
does not hold, for while a few of the largest specimens measure in 
length of head and body, 203, 215, and even 231 mm., most of them are 
below 200 mm. given for the type of ankoliae. 

Measurements, cf . 231. 55. 27. 9 mm., 9-210. 73. 30. 29 mm., both 
from Nyakabande. 

Parasites. Fleas ( ) were present on a Mush- 

ongero rat. 



MURIDAE 

Dendromus insignis kivu Thomas 

Dendromus insignis kivu Thomas, 1916, Ann. Mag. Nat. Hist. (8), 12, p. 242: 
Buhamba, Kivu region, Belgian Congo. 

7 c? 15 ? 2 juv. (M. C. Z. 40767-90) Idjwi Id., B. C. 20. ii-3. iii. 39. 

Native name. Shungwe (Lulega). 

Discussion. Though but slightly marked, this seems to be a valid 
race with a distinctly narrower dorsal stripe, less broadly expanded 
over the shoulders than in the typical form, and with the under surface 
on the average slightly huffier. In size, however, there is probably 
little, if any, difference when sufficient series are compared. Of the 
present series, hardly half are adults. 

Measurements, d". 85. 102. 17. 14 mm., 9 . 87. 95. 20. 13 mm. 

Breeding. On February 21, an 82 mm. mother and her four young 
(of which a 6^ measured 47. 46. 13. 6 mm.) were brought in, the latter 
squeaking noisily though their eyes were unopened. The following day 



ALLEN AND LOVERIDGE: AFRICAN MAMMALS 189 

an 85 mm. mother and three young (cf. 61. 77. 18. 10 mm., 9 . 62. 82. 
18. 10 mm.) arrived, and several other htters on succeeding days. 

Enemies. One recovered from the stomach of a snake (Boaedon I. 
lineatus). 

Dendromus messorius ruddi Wroughton 

Plate 4, fig. 1. 

Dendromus ruddi Wroughton, 1910, Ann. Mag. Nat. Hist. (8), 5, p. 275: 
Malikisi, Mt. Elgon, Kenya Colony. 

5 d' 10 9 (M. C. Z. 39334-6, -8-42, 40812-8) Bundibugyo, U. 20- 
24. xii. 38. 
1 9 (M. C. Z. 39337) Mihunga, Ruwenzori Mtns., U. 14. i. 39. 

Native names. Kamampi (Luamba); kuinji (Lukonja); mbeba wa 
irungu (Lutoro). 

Discussion. This series of sixteen specimens is of uniform appearance, 
long-tailed and white-beUied, hind foot about 18 mm. They agree so 
closely in proportions and in the rich tawny color of the back with 
D. messorius of the Cameroons, that there can be no doubt of their 
close relationship. The latter is slightly more richly colored with an 
ochraceous wash over the belly. Dr. R. T. Hatt (1940, p. 482), in- 
cludes specimens from Medje and Niangara, eastern Congo, under 
typical messorius, which therefore ranges quite across the western and 
central forest area. 

Measurements, cf . 72. 91. 17. 12 mm., 9 . 77. 93. 17. 12 mm. 

Breeding. On January 14, at Mihunga, a 65 mm. female was found 
with her nest, constructed of soft coarse grasses and lined with soft 
fine grasses and grass heads, measuring about 80 mm. in diameter. It 
contained three naked, blind nestlings measuring 35. 16. 6. 2.5 mm. 
The nest and young were subsequently photographed in a domestic 
banana, in which plant the nests are found most frequently. 

On January 23, at Bugoye, eastern foot of Ruwenzori, a native 
brought in four young with their eyes open. We released these in 
southwestern Kigezi the following week. 

Dendromus whytei pallescens Osgood 

Dendromus whytei pallescens Osgood, 1910, Fubl. Field Mus. Nat. Hist., Zool. 
Ser., 10, p. 7: Lukenya, Ulukenya Hills, Kenya Colony. 

9 (M. C. Z. 38845) Mbanja, T. T. 29. iv. 39. 

9 (M. C. Z. 39059) Magrotto Mtn., T. T. 15. vii. 39. 



190 bulletin: museum of comparative zoology 

Native names. Nkule (Kimakonde, but not even generic); daa 
(Kisambara). 

Discussion. The Magrotto specimen is slightly paler than typical 
examples and has a faint indication of a dark dorsal stripe on the rump. 
The tail is shorter than in D. m. ruddi: 77 and 72 mm. respectively in 
these two females. 

Measurements. 9 • 65. 77. 14. 11 mm., from Mbanja. 

Breeding. On April 29, this 65 mm. female was brought in with a 
nest composed of fine grasses and a concealed elastic opening. The 
structure measured approximately 115x65 mm. (41^x23^ inches) 
and held seven naked nestlings. On July 15, a 62 mm. female had only 
four young, these were furred and their eyes already opened. 

Ca-ptivity. As Magrotto Mountain, at 3000 feet, was quite a tem- 
perate climate, overclouded and with heavy precipitation during much 
of our stay, I attempted to take the four above-mentioned nestlings 
back to Europe. In the Red Sea, however, it was very hot— said to be 
140° in the sun while we were at Port Sudan— and two of the mice suc- 
cumbed. The remaining two, after staging brief escapes in the Paris- 
Bologne and London-Cardiff express, reached Glamorganshire safely. 
Cold weather apparently affected them but little, for my niece, to 
whom they were consigned, informs me that they passed the coldest 
fortnight of the exceptionally severe winter of 1939 in an unheated 
stone outhouse in Devonshire. However, during a cold spell in Decem- 
ber, 1940, they both died one night without any other apparent cause. 

In an attempt to tame them, they were handled as much as possible. 
To do so, however, it was necessary to entice them from their nest. 
This was resented at times, one used its teeth in protest, and as if seek- 
ing freedom from molestation, they built a new nest. 

Their roomy cage held several branches, up and down which the 
active little creatures enjoyed exercising, balancing on the most 
slender terminal twigs, or hanging upside down from the roof of their 
cage while they tore off from it splinters of wood. 

The food which was furnished them consisted of milk, honey, grapes, 
bananas, apples, monkey nuts, biscuits, crushed oats, sundry scraps, 
but principally parrot and canary seed, quantities of which they con- 
sumed, though all these foods appeared to be appreciated. 

Thamnomys sifRDASTER suRDASTER Thomas & Wroughton 
Thamnonujs surdaster Thomas & Wroughton, 1908, Froc. Zool. Soc. London, 
p. 550: Zomba, Nyasaland. 

d^ 9 (M. C. Z. 39054-5) Magrotto Mtn., T. T. 5. vii. 39. 



ALLEN AND LOVERIDGE: AFRICAN MAMMALS 191 

Native name. Kozo (Kisambara). 

Discussion. These specimens come from a point within twenty 
miles of the type locality of T. usamharae, apparently a synonym of 
surdaster. It might have been thought that coastal skins would have 
been identical also with the race Uttoralis, described from Mazeras, 
Kenya Colony, but the latter have nearly pure white feet, whereas in 
the Magrotto mice the metatarsal region of the feet is buffy. 

Measurements, d" . 117. 165. 24. 16 mm., 9 • 110. 145. 24. 12 mm. 

Enemies. A halfgrown mouse was recovered from the stomach of a 
snake (Boaedon I. lineatus). 

Thamnomys SURDASTER DRYAS Thomas 

Thamnomys dryas Thomas, 1907, Ann. Mag. Nat. Hist. (7), 19, p. 123: 
Mubuku Valley, Ruwenzori Mountains, Uganda, 6000-7000 ft. 

2 9 (M. C. Z. 39202-3) Mihunga, U. 29. xii. 38-16. i. 39. 
2 cf^ 1 9 (M. C. Z. 39247-9) Idjwi Id., B. C. 20. ii-3. iii. 39. 

Native name. Misisi (Lulega). 

Discussion. The Mihunga specimens are absolute topotypes, those 
from Lake Kivu, being somewhat immature, are doubtfully identical. 
They differ from typical surdaster in their slightly smaller (21 mm.) 
hind feet, their backs are a trifle darker, their skulls slightly smaller, 
which, however, may be due to immaturity, but the hindmost cusp of 
the last upper molar shows no distinct division into two by a median 
notch at the hinder end as is the condition in the two typical surdaster 
from Magrotto. As suggested by Dr. R. T. Hatt (1940, p. 539) in the 
case of one from the same locality, they perhaps represent intermedi- 
ates between T. s. dryas and T. s. surdaster. In the former the pectoral 
pair of mammae is said to be absent, but this may vary individually. 
Nor is it clear that all are not best considered as races of T. rutilmis 
of West Africa. 

Measurements. 9 . 123. 142. 22. 16 mm., from Mihunga, Ruwen- 
zori. ■d'. 91. 114. 20. 10' mm., 9- 92. 132. 21. 15 mm., both from 
Idjwi Id. 

Thamnomys venustus venustus Thomas 

Thamnomys venustus Thomas, 1907, Ann. Mag. Nat. Hist. (7), 19, p. 122: 
Mubuku Valley, Euwenzori Mountains, Uganda, ^6000 ft. 

cf (M. C. Z. 39201) Mihunga swamp, U. 18. i. 39. 

Discussion. This again is topotypical. 
Measurements, cf . 155. 205. 25. 23 mm. 

' Given as 6000 ft. in original, 8000 ft. in Trans. Zool. Soc. London, 1919, p. 508. 



192 bulletin: museum of comparative zoology 



Thamnomys venustus kivuensis subsp. nov. 

Type. Museum of Comparative Zoology, No. 39151, an adult 
female, skin and skull, from Upper Mulinga, Idjwi Island, Lake Kivu, 
Belgian Congo, collected by Arthur Loveridge, March 2, 1939. 

Description. Similar in size, proportions, and tooth characters to 
typical venustus from Ruwenzori, but with the prevailing tint of the 
dorsal surfaces of the body much duller, warm buff to ochraceous buff 
instead of 'ochraceous tawny' as in the latter. 

Pelage long and silky, hairs of the back about 17 mm. long, slaty 
in their basal four-fifths, tipped with ochraceous bulT and mixed with 
scattered all-black hairs. Sides of the face and upper part of forearms 
clearer, more intense ochraceous; flanks paler than back, with few 
black hairs, the color passing gradually into that of the ventral side, 
which from chin to anus is dull whitish, faintly washed with buify at 
the tips of the hairs, the basal half or so slaty; anal region clearer 
ochraceous buff. Ears nearly naked, their substance pigmented a dark 
brown, with scattered short brown hairs externally, ochraceous in- 
ternally. Fore feet with metacarpal area dark brown, toes whitish; 
hind feet ochraceous buff, clear, except at the bases of the toes, which 
are clouded with dusky. Tail much longer than head and body, dark 
blackish brown all around, the hair of the basal two-thirds short, dark 
and close, becoming longer in the terminal third to form a thin tuft. 

Skull quite like that of T. venustus, with the postero-internal cusp 
of all three upper molars well developed and as large as the two in 
front of it. Each upper molar, therefore, has three sub-equal elongate 
cusps on the lingual side, while on the labial side molars 1 and 2 have 
each four small cusps of which the posteriormost is the smallest. 

Measurements. The collector's measurements are : head and body, 
145 mm.; tail, 173; hind foot (s. u.), 26; ear, 18. The skull is crushed 
but shows the following: upper molar row, 6 mm.; diastema, 8.8; 
width outside first molars, 6.1; breadth of brain case, 13.8; mandible 
from condyle to tip of incisor, 19.6. 

Remarks. The tree rats of this group are difficult to secure; seldom 
more than one or two occur in even larger collections, while the smaller 
forms, such as T. surdaster, seem commoner or at least easier to trap. 
Although no others were secured in the Kivu region, the duller, less 
intense coloring of the single specimen as compared with those from 
Ruwenzori is paralleled by a similar less-intense coloring in the repre- 
sentative forms of Dendromus insignis and Leggada bufo of these areas, 
and hence is probably significant. The Museum of Comparative Zool- 



ALLEN AND LOVERIDGE: AFRICAN MAMMALS 193 

ogy has an example of typical vcnustus from Kibati, southeastern base 
of Mt. Niragongo, Kivu volcanoes, so that its distribution probably 
extends from Ruwenzori south to this range, while still farther south- 
ward beyond these mountains the duller race here described is found. 
Habits. At dusk each evening I observed this animal descending a 
vine in the dense tangle of secondary forest immediately behind my 
tent. Baiting a snap-back rat trap with a scrap of pawpaw fruit, I 
set it on the vine at a height of about six feet from the ground and was 
successful in securing the rat. 

Oenomys hypoxanthus unyori (Thomas) 

Mus hypoxanthus rinyori Thomas, 1903, Ann. Mag. Nat. Hist. (7), 12, p. 342: 
'Fadjas, on Victoria Nile, Unyoro, Uganda. 

9 (M. C. Z. 39289) Bundibugyo, U. 20. xii. 38. 

Native name. Ndoga (Luamba); nsisa (Lutoro). 

Discussion. In this specimen the midventral area is pure white 
with a buffy line along the lateral border. The distinction between 
this form and Oe. h. bacchante, described from Xandi, Kenya Colony, 
does not seem well founded, and Hollister doubts if editus is really dis- 
tinguishable from unyori. 

Measurements. 9 . 140. 180. 27. 19 mm. 

Oenomys hypoxanthus editus Thomas & Wroughton 

Oenomys bacchante editus Thomas & Wroughton, 1910, Trans. Zool. See 
London, 19, p. 509: Mubuku Valley, Ruwenzori Mountains, Uganda, 
6000 ft. 

cT (M. C. Z. 40830) Mihunga, Ruwenzori Mtns., U. 14. i. 39. 
cf (M. C. Z. 39144) Kisenyi, Lake Kivu, B. R. 10. ii. 89. 
3 9 (M. C. Z. 39142-3, 40744) Idjwi Id., B. C. 20-23. ii. 39. 

Native names. Mbiitu (Lukonjo); nsisa (Lutoro). 

Discussion. The Lake Kivu specimens agree closely with the topo- 
typical Ruwenzori rat representing the race editus, in having the under 
surface of the body heavily washed with ochraceous, except for the 
axillae and groins, which are white. In typical Oe. h. hypoxanthus, 
from the Cameroons, the lower surface is tinted faintly, or not at all, 
with pale ochraceous while the fore part of the body above is more 
coarsely speckled with the same. 

Measurements, d". 135. 140. 28. 18 mm., 9 . 170. 200. 33. 21 mm., 
from Kisenyi and Idjwi, respectively. 

' Fajao, Murchison Falls, Victoria Nile. 



194 bulletin: museum of comparative zoology 

Breeding. On February 23, a 166 mm. female was taken with her 
three blind nestlings, one of which (M. C. Z. 40744), a 9 , measured 
70. 50. 17. 7 mm. 

Diet. Two were trapped with bread bait, one by its tail; unfortun- 
ately a sixth was destroyed by some other rat eating out its brains as 
it lay dead in the trap. 

Rattus rattus kijabius (Allen) 

Mus kijabius J. A. Allen, 1909, Bull. Amer. Mus. Nat. Hist., 26, p. 169: 
Kijabe, Kenya Colony. 

1 c? 1 9 (M. C. Z. 39208, 40837) Isungo, near Kibale Forest, U. 14. 
xii. 38. 

1 9 (M. C. Z. 39290) Bundibugyo, Bwamba region, U. 20. xii. 

38. 

1 cf 2 9 (M. C. Z. 40834-6) Mihunga, Ruwenzori Mtns., U. 4. i. 39. 

2 9 (M. C. Z. 39209-10) Nyakabande, Kigezi, U. 28. i. 39. 

2 c? 1 9 (M. C. Z. 39155-7) Idjwi Id., Lake Kivu, B. R. 17. ii. 39. 

1 d' (M. C. Z. 38873) Kitaya, Ruvuma River, T. T. 30. iii. 39. 

Seen also at Mushongero, Kisenyi, Ujiji, and Magrotto Mountain. 

Native names. Mpanya^ (Lukonjo and Lutoro) ; mbeba (Lukiga and 
Lulega); nkule (Kimakonde); likosive (Kiyao); ngoshwe (Kisambara). 

Discussion. This is the common house rat of eastern Africa, abun- 
dant in all the native villages, but whether or not it is an introduction 
from India, or a local race indigenous to this part of Africa, is at 
present obscure, though the latter alternative seems likely. Immature 
animals have the entire lower surface slaty gray, often with a faint 
tinge of bufTy ; the lower side has more buff-ringed hairs, producing in 
adults a mixed black and buff effect. 

Measurements, d". 155. 185. 31. 21 mm., 9 . 180. 213. 32. 23 mm., 
from Idjwi and Mihunga, respectively. 

Diet. Feeding in a mahoga garden at Mihunga. 

Enemies. A full-grown rat occupied the entire stomach of an owl 
{Bubo a. africanus) shot at Magrotto. A very large male, measuring 
350 (190 -f 160) mm., and two half-grown rats, in the stomachs of 
house snakes {Boaedon I. Rneatus) at Ujiji, Fort Portal and Magrotto; 
four very large ones in Gaboon vipers {Bitis gabonica) at Budongo and 
Magrotto, three in nose-horned vipers (B. yiasicornis) at Mabira and 
on Idjwi Island. Four furred nestlings in a cobra {Naja n. nigricollis) 
killed in an Indian store at Kitaya. 

' This is the Kiswahili name for this recent immigrant; the Uganda Education Department 
informs me that Nsolima is more correct for Lukonjo and Lutoro. 



ALLEN AND LOVERIDGE: AFRICAN MAMMALS 195 

Remarks. Woosnam remarked that the Ruwenzori Expedition 
failed to find this rat on the mountain in 1905-6. Now, in 1939, it is 
plentiful about the native huts in the Mubuku Valley at 6000 feet, 
and it remains to be seen what effect it will have on the numerous 
kinds of smaller rodents occurring in the \acinity. 

Aethomys kaiseri hindei (Thomas) 

Mus hindei Thomas, 1902, Ann. Mag. Nat. Hist. (7), 9, p. 219: Machakos, 
Kenya Colony. 

2 d^ 3 9 (M. C. Z. 39056, -132-4, -6) Amboni Estate, T. T. 19. vi. 

39. 
1 cT 6 9 (M. C. Z. 39126-31, -5) Magrotto Mountain, T. T. 15-17. 

vii. 39. 

Native name. Sase (Kisambara). 

Discussion. On account of its shorter pelage, as compared with 
Ae. k. norae of northern Kenya Colony, the black hairs scattered num- 
erously throughout the dorsal pelage produce less of a lined than a 
minutely peppered effect. The relatively shorter tails are, as Hollister 
pointed out, an obvious characteristic. Based on field measurements, 
the tail averages 83% of the head-and-body length (extremes 71 to 90). 

Measuremenls. d". 163. 117. 29. ?20 mm., 9 . 175. 153. 23.20 mm., 
both from Magrotto. 

Breeding. On June 19, a 155 mm. female, dislodged by a tractor 
engaged in spreading piles of decaying vegetation, ran from it dragging 
after her three large young, attached to her nipples. One of these, a cf , 
measuring 70. 50. 15. 8 mm., was made into a skin (M. C. Z. 39056), 
the two others were preserved in alcohol. I had been told previously 
that for females to carry their young in this fashion was a common 
practice of rats in this neighbourhood On the peculiar forked tip of 
the incisors in the young, see Lawrence (1941.) 

Enemies. The native youngster who brought me the big series on 
March 17, was surprised and pleased at my buying the lot, exclaiming: 
"I have erred in leaving four at my house. Do you want them too?" 
I replied in the affirmative, provided that they were undamaged. "I 
will fetch them at once, but it is a long way," said he, departing in 
haste. Later, he found me frogging in a marsh and proffered his shirt- 
skirt full of fowl's eggs. "But where are the rats?" I asked. Instead of 
replying to me in Kiswahili, he spoke rapidly in Kisambara to my local 
gunbearer, as if uncertain whether it would be good form to tell me the 
truth! "What does he say?" I enquired. "He says," repHed the man. 



196 bulletin: museum of comparative zoology 

"that when he got home he found the children had toasted them, 
skins and all, and eaten them." Lest anyone, with different tastes, 
should suppose that this reflected poverty, I might add that the pican- 
ninies on Magrotto Estate were exceptionally plump through feeding 
on the oil-palm nuts along with the civets, squirrels, rats, vultures, 
crows and other creatures. 

Praomys jacksoni MONTIS (Thomas & Wroughton) 

Af MS jacksoni montis Thomas & Wroughton, 1910, Trans. Zool. Soc. London, 
19, p. 503: Mubuku Valley, Ruwenzori Mountains, Uganda, 6000 ft. 

2 9 (M. C. Z. 39232, 39329) Kibale Forest, U. 10. xii. 38. 
1 & (M. C. Z. 40797) Bundibugyo. U. 20. xii. 38. 

3 d^ 1 9 (M. C. Z. 39235-6, 39326, 40819) Mubuku V., U. 3-6. i. 39. 
2 d' 6 9 (M. C. Z. 39233-4, -7-8, 39327, -30-32) Mihunga, U. 14- 

22. i. 39. 

4 cf 4 9 (M. C. Z. 39150, 39240-44, -46, 40734) Idjwi Id., B. C. 

24. ii-2. iii. 39. 

Native names. Mhule (Lukonja); udiakiru (hutoro) ; sisisi (Lulega). 

Discussion. The series from Idjwi Island seems identical with the 
topotypes from Ruwenzori. One large 135 mm. (head and body) fe- 
male has a skull with condylobasal length of 30.5 mm. The tipping of 
the dorsal hairs in adults is pale ochraceous instead of the brighter 
cinnamon of typical jacksoni. Two from Kibale forest are in the bright 
cinnamon pelage, rather brighter in tint than M.C.Z, 39232, the only 
one of the Ruwenzori series in this phase. 

Measurements. &. 160. 172. 30. 18 mm., 9. 135. 160. 26. 17 mm., 
from Mihunga and Idjwi, respectively. 

Diet. Trapped with cheese. 

Breeding. On February 24, the large female, mentioned above, was 
brought in together with two, still blind, nestling males measuring 68. 
62. 17. 9 mm. 

Enemies. Six two-thirds grown young in the stomach of a mamba 
{DendrocLS-pis j. kaimosae). 

Praomys taitae (Heller) 

Epimys taitae Heller, 1912, Smithsonian Misc. Coll., 59, No. 16, p. 9: Mt- 
Mbololo, Teita Hills, Kenya Colony, 5000 feet. 

a" (M. C. Z. 39057) Magrotto Mtn., T. T. 11. vii. 39. 
Native name. Hunju (Kisambara). 



ALLEN AND LOVERIDGE: AFRICAN MAMMALS 197 

Discussion. Compared with topotypes obtained by Loveridge in 
1934. It is interesting to note that the present specimen was obtained 
under closely similar habitat conditions. 

Measurements, cf . 102. 111. 23. 18 mm. 

Habitat. Noticing a burrow between the buttress roots of a huge 
tree in the heart of the forest, I ordered the gunbearers to dig. At 
eighteen inches or thereabouts below the surface the burrow, which had 
sloped steeply downwards to this point, turned upwards and divided 
into two short passages. One held a loose assemblage of dry leaves in a 
fragmentary state in which this rat was hiding, the other was a blind 
passage. 

Hylomyscus denniae denniae (Thomas) 

Musdenniae Thomas, 1906, Ann. Mag. Nat. Hist. (7), 18, p. 144: Mubuku 
Valley, Ruwenzori Mountains, Uganda, 7000 feet. 

d^ (M. C. Z. 39328) Mubuku Valley, U. 6. i. 39. 

Discussion. This single topotype, also taken at 7000 feet, was the 
only one obtained by Loveridge during his week at Mubuku camp, 
though R. B. Woosnam found it "extremely common" at the same spot 
in 1906. 

Hylomyscus carillus schoutedeni (Dollman) 

Epimys schoutedeni Dollman, 1914, Revue Zool. Africaine, 4, p. 82: Mambaka, 
Belgian Congo. 

2 9 (M. C. Z. 39239, 39245) Idjwi Id., B. C. 25. ii. 39. 

Native name. Mtumhabuva (Lulega). 

Discussion. An adult female and its two-thirds grown young are 
referred to the form schoutedeni, which Dr. Hatt (1940, p. 537) re- 
gards as a race of carillus. It is much brighter tawny above than H. d. 
denniae and its races, or than H. a. weileri, has 1 — 2 = 6 mammae, a 
tail slightly pencilled, hind feet with a slight clouding of darker at the 
base of the toes but without dark tarsal mark. The edges of the frontal 
are strongly beaded, with a minutely projecting point just in advance 
of the parietal. The immature specimen is dark, with only a sprinkling 
of minute tawny tips, especially on the head, and very dark metatarsal 
area and white toes. This identification, if correct, carries the range 
eastward to the extreme edge of the Congo region. 

Measurements. 9 . 100. 137. 19. 16 mm., 9 yng. 80. 100. 17. 14 mm. 

Breeding. Taken from a large, loose nest constructed of strips of dry 
banana-leaf, built in a bunch of wild banana fruit that was hanging 
upside down. A second young one escaped. 



198 bulletin: museum of comparative zoology 

Mastomys coucha ugandae (De Winton) 

Mus ugandae De Winton, 1897, Ann. Mag. Nat. Hist. (6), 20, p. 317: Entebbe, 
Uganda. 

9 (M. C. Z. 40831) Isunga, near Kibale Forest, U. 14. xii. 38. 
5 cf 3 9 (M. C. Z. 39319-25, 39333) Bundibugyo, U. 20. xii. 38. 

Native navie. Bandugi (Luamba); ndiakiru (Lutoro). 
Measurements. &. 155. 87 + . 25. 17 mm., 9 • 145. 120. 27. 18 mm. 
Both from Bundibugyo. 

Mastomys coucha durumae (Heller) 

Epimys coucha durumae Heller, 1912, Smithsonian Misc. Coll., 59, No. 16, 
p. 9; Mazeras, Kenya Colony. 

4 c^ 3 9 (M. C. Z. 39120-5, 39137) Magrotto Mtn., T. T. 10-17. 
vii. 39. 

Native name. Shishe (Kisambara). 

Discussion. This is a poorly marked race of the hot coastal strip 
of southeastern Kenya Colony, ranging into the adjacent Tanganyika 
Territory for an undetermined distance. It is slightly grayer and less 
brownish than the earlier-described hildebrandtii of the Teita Hills 
region. Externally it much resembles immature examples of the race of 
Aethomys kaiseri occurring in the same region, but is easily distin- 
guished by its shorter and more slender foot, about 25 mm. with claw, 
and by its somewhat closer pelage. The longer, slit-like incisive fora- 
mina at once distinguish the skull. 

Measurements, cf . 145. 126. 22. 19 mm., 9 . 120. 105. 22. 18 mm. 

Enemies. Three rats, apparently of this species, recovered from the 
stomach of a viper (Bitis gabonica) killed close to the spot where this 
series was trapped. 

Habitat. Trapped at the edge of a swamp in a valley on the moun- 
tain. Two others trapped, had their skulls eaten out by other rats. 

Mastomys coucha microdon (Peters) 

Mus microdon Peters, 1852, Reise nach Mossambique, Si ugeth., p. 149, pi. 
XXXV, figs. 5-6, pi. xxxvi, fig. 1 : Tette & Boror, Mozambique. 

d^ (M. C. Z. 38841) Kitaya, T. T. 28. iii. 39. 
9 (M. C. Z. 38874) Mbanja, T. T. 27. iv. 39. 

Native names. Chikukumula (Kimakonde at Kitaya) ; nkule (Kim- 
akonde at Mbanja). 

Discussion. A somewhat paler race than M. c. ugandae. 



ALLEN AND LOVERIDGE: AFRICAN MAMMALS 199 

Measurements, cf. 140. 110. 24. 16 mm., 9 . 122. 107. 22. 18 mm. 

Breeding. On April 27, the female held fifteen small embryos (pre- 
served). 

Habitat. The male was taken in a depression, lined with a few 
grasses, beneath a pile of weeds at the edge of a rice swamp — an 
extremely damp situation. 



Leggada bufo bufo Thomas 

Leggada bufo Thomas, 1906, Ann. Mag. Nat. Hist. (7), 18, p. 145: Mubuku 
Valley, Ruwenzori Mountains, Uganda, 6000 feet. 

2 cT 4 9 (M. C. Z. 40820-2, -4-5, -8) Mihunga, U. 12-18. i. 39. 

Native name. Kienje (Lukonjo). 

Discussion. These small orange-bellied mice are topotypes. In 
youngish specimens the color of the belly is much duller than in the 
adults, gray with a wash of ochraceous. 

Measurements, d". 85. 61. 15. 11 mm., 9 • 85. 70. 16. 12 mm. 



Leggada bufo ablutus subsp. nov. 

Type. Museum of Comparative Zoology, Xo. 40745, an adult male, 
skin and skull, from Upper Mulinga, Idjwi Island, Lake Kivu, Belgian 
Congo, collected by Arthur Loveridge, February 24, 1939. 

Description. Like typical L. bufo of Ruwenzori, but distinguished 
from it by its slightly less-dark dorsal coloring and by its paler under 
surface in which the gray-based hairs from chin to root of tail have a 
subterminal ring of whitish and a tip of bright ochraceous buff, pro- 
ducing a wash of this tint rather than the richer 'ochraceous orange' 
of bufo. 

The general color above is a uniform mixture of black, finely punc- 
tate with orange, hardly darker in the center of the back, but slightly 
less dark on cheeks and forehead; no subauricular spot and no eye- 
ring; ears blackish brown, nearly naked; tail blackish above, paler 
below, minutely haired. Fore feet pale or slightly darkened on the 
metacarpal area; hind feet similar. 

The skull does not differ from that of typical bufo. Upper incisors 
very slightly notched; incisive foramina long, extending back to about 
the anterior third of the first molar; masseteric knob prominent. 
First upper molar with three outer and two inner cusps, of which the 



200 bulletin: museum of comparative zoology 

antero-external one is small ; the antero-internal cusp much behind the 
transverse level of the former. 

Measurements. The collector's measurements of the type and an 
adult female para type, respectively, are: head and body, 90, 86 mm.; 
tail, 70, 75; hind foot, 16, 15; ear, 11, 12. The skull of the type meas- 
ures: greatest length, 22.0 mm.; basal length, 17.6; palatal length, 
10.4; zygomatic width, 10.0; mastoid width, 9.2; width across molars, 
4.9. 

Remarks. The three adult specimens (M. C. Z. 40745-7) from Lake 
Kivu are so noticeably paler below than the topotypes of L, h. bufo 
that they seem to represent a local form worth recognition. A similar 
paling out in color is seen in some other local representatives of small 
mammals as compared with Ruwenzori specimens, as for example 
Dendromus insignis kivu. 

Native name. Muhushushu (Lulega). Despite the similarity in the 
Lulega names for this small rodent and for the shrews, I was assured 
that the difference in pronunciation had been correctly transcribed. 

Leggada grata grata Thomas & Wroughton 

Leggada grata Thomas & Wroughton, 1910, Trans. Zool. Soc. London, 19, 
p. 507: Mubuku Valley, Euwenzori Mountains, Uganda, 6000 ft. 

9 (M. C. Z. 40826) Mihunga, U. 12. i. 39. 
9 (M. C. Z. 40827) Nyakabande, U. 2. i. 39. 
6 c^ 9 9 (M. C. Z. 40748-62) Idjwi Id., B. C. 18. ii-3. iii. 39. 

Native name. Mpiongo (Lulega). 

Discussion. The specimen from Mihunga, Ruwenzori, is topotypic 
of this species, and though slightly darker above than the Kivu series, 
perhaps owes this difference to its immaturity. In immature speci- 
mens the bufFy line at the edge of the belly is not developed. 

Measurements, cf. 78. 57. 13. 10 mm., 9- 75. 55. 12. 9 mm., both 
from Idjwi. 

Breeding. On February 20, a 75 mm. female was found in her nest 
of dry grass beneath a heap of garden rubbish. The four young present 
had their eyes closed, and consisted of a cf. 43. 33. 11. 5 mm., and 
three females 43. 35. 11. 5 mm. (preserved). 

Enemies. Two recovered from the stomach of a mamba (Dendro- 
aspis j. kaimosae) on Idjwi, and two from vipers {Atheris nitschei) on 
Ruwenzori and Idjwi, respectively. 



ALLEN AND LOVERIDGE: AFRICAN MAMMALS 201 



Leggada BELLA viciNA Thomas 

Leggada hella vicina Thomas, 1910, Ann. Mag. Nat. Hist. (8), 6, p. 88: Tak- 
aungu, near Mombasa, Kenya Colony. 

1 cf 2 9 (M. C. Z. 38861, 38864-5) Kitaya, Rovuma R., T. T. 
29. iu. 39. 

1 d^ 3 9 (M. C. Z. 38847-8, 38862-3) Miklndani, T. T. 17-19. 
iv. 39. 

1 cf 1 9 (M. C. Z. 38846, 38880) Mbanja, near Lindi, T. T. 27. 
iv. 39. 

2 d' (M. C. Z. 38866-7) Mainland opp. Ivilindini, K. C. 25. ' 
vii. 39. 

Native names. Chanile (Kimakonde at Kitaya) ; ngorpo (Kimakonde 
at Mbanja); irutu (Kimwera). 

Discussion. No difference appears between topotypical skins from 
near Mombasa and those of the southeastern part of Tanganyika. 
There is some individual variation in the amount of darkening of the 
back and the intensity of the ochraceous on the sides, partly it seems, 
a matter of age. 

Measurements, d". 68. 42. 12. 11 mm., 9 . 68. 46. 12. 10 mm., from 
opposite Kilindini, and Kitaya, respectively. 

Breeding. On March 29, two nests, comprised of a loose assemblage 
of fine grasses without any lining, were found on the ground beneath 
piles of (i) thatching grass, (ii) weeds. In addition to the mother mice, 
they held litters (preserved) consisting of blind and naked nestlings of 
a very raw red hue, and numbering eight and three respectively. 

On April 17, a loosely built nest measuring about 100 mm. (circa 4 
inches) in diameter, constructed of finely shredded outer leaves of 
maize, was found on the ground beneath a pile of discarded corncobs 
and their husks. In addition to the 66 mm. mother, it held seven 
naked nestlings (preserved), one of which measured circa 37. 24. 6. 
3 mm. 

On April 19, a 67 mm. female held seven fetuses (preserved). 

On April 27, a litter consisting of six naked nestlings (preserved) 
was found. 

Diet. Finely masticated maize found in stomach of one mouse. 

Enemies. An adult in the stomach of a snake (Boaedon I. lineatus) 
at Kitaya; three large and one small nestling in a burrowing viper 
(Atractaspis hihronii) at Mbanja. 



202 bulletin: museum of comparative zoology 

Cricetomys gambianus proparator Wroughton 

Cricetomys gambianus proparator Wroughton, 1910, Ann. Mag. Nat. Hist. (8) 
5, p. 107: Mubuku Valley, Ruwenzori Mountains, Uganda, 6000 ft. 

1 9 (M. C. Z. 40792) Mabira Forest, U. 16. xi. 38. 
3 cf (M. C. Z. 39420-2) Bundibugyo, U. 20. xii. 38. 
1 9 (M. C. Z. 39419) Mihunga, U. 13. i. 39. 

Native names. Kayozi (Luganda); msumha (Luamba and Lutoro); 
isisa not chiclia (Lukonjo). 

Discussion. The specimen from Mihunga is topotypical of prop- 
arator, dark above, white below, the fore feet white-toed, the hind feet 
dark with paler toes, tail dark in its basal third. The specimen from 
Mabira is practically identical with it but has the chest slightly clouded 
with gray. The three Bundibugyo specimens are slightly more buffy 
than proparator, and are not very different from topotypes of elgonis. 
In one the under side is all white, in another the chest is marked with 
gray, while in the third the entire lower surface is gray. In their fore 
feet the condition varies individually from white-toed to having the 
entire foot white, the hind toes are white and the tails about one-half 
white. 

Dr. R. T. Hatt (1940, p. 493) has remarked at length on the varia- 
tion shown by these giant rats, especially in the regions where forest 
and savannah habitats meet or overlap, and on the difhculty of assign- 
ing individual specimens to a given race. Without sufficiently repre- 
sentative series it is almost impossible to find sharply delimiting char- 
acters for local races. Nevertheless he finds those of the Congo forest 
rather sharply marked off from those of the eastern savannah, so that 
he regards them as specifically distinct under the name of C. dissimilis 
(misspelled dissimulus). Probably the explanation is that the savannah 
forms with their clouded bellies and paler, more ochraceous pelage, 
mingle with the white-bellied, darker-backed forest forms along the 
eastern outposts of the central forest area, especially where the latter 
is gradually being driven back so that on the borders of the two habit- 
ats the differences become less clear, and the characters more or less 
intermediate. The case is perhaps somewhat paralleled by those of the 
forest and savannah elephants, the dwarf and Cape buffaloes, and 
others, which seem to be pairs of forms living typically either in the 
forest or in the open country, but due to lack of any but environmental 
barriers are not yet completely segregated as distinct species. It 
becomes then a matter of opinion whether to regard them as distinct 
species or as representative forms of a single species. 



ALLEN AND LOVERIDGE: AFRICAN MAMMALS 203 

Measurements. &. 390. 430. 65. 46 mm., 9 . 320. 370. 71.42 mm., 
from Bundibugyo and Mihunga, respectively. 

Parasites. Hemimerids ( ) in fur of the 

Mabira and Bundibugyo rats, but only a tick (Ixodes rasus) on the 
Mihunga specimen. 

Cricetomys gambianus osgoodi Heller 

Cricetomys gambianus osgoodi Heller, 1912, Smithsonian Misc. Coll., 59, No. 
16, p. 16: Mazeras, Kenya Colony. 

& (M. C. Z. 39423) Nchingidi, T. T. 19. v. 39. 

Discussion. This specimen agrees precisely with the characters 
pointed out by Heller. The mesopterygoid fossa of the skull is narrow 
and parallel-sided, the bullae are small, and the zygomata are obviously 
more bowed than in the Uganda specimens; skull length, 72 mm. 
Compared with the races proparator and elgonis, the dorsal coloring is 
huffier, belly clear white, fore feet pale, no light area at the anterior 
base of the ear. 

LoPHUROMYs AQUILUS AQUILUS (True) 

Mus aquilus True, 1892, Proc. U. S. Nat. Mus., 16, p. 460, fig. 1 : Mt. ICiliman- 
jaro, Tanganyika Territory, 8000 feet. 

1 cf 2 9 (M. C. Z. 39061-3) Magrotto Mtn., T. T. 14-16. vii. 39. 

Native name. Vusu (Kisambara). 

Discussion. In a previous report (1936, p. 99), with an extensive 
series of material from Mt. Elgon, it was considered that rubecula 
DoUman, described from Elgonyi, Mt. Elgon, was unrecognizable from 
the nominate form. The three specimens listed above are slightly less 
dark above, redder on the sides, and pinker on the belly than those 
from Ruwenzori and Lake Kivu listed below. 

Measurements, d". 145. 80. 21. 18 mm., 9 . 144. ?. 22. 18 mm. 

Breeding. On July 14, a female held four very small embryos 
(preserved). 

Diet. Palm-oil nut and other matter in one stomach. 

Enemies. No parasites, but note loss of tail by both females! 

LoPHUROMYS AQUILUS (subsp.) 

cf (M. C. Z. 39223) Budongo Forest, U. 28. xi. 38. 



204 bulletin: museum of comparative zoology 

Discussion. This skin is almost without darkening of black hairs 
dorsally, but is evenly punctate with black and ochraceous, paler even 
than the race zena, while the belly is pure pinkish ochraceous. 

Measurements, cf . 135. 65. 19. 16 mm. 

Diet. Trapped with bread as bait. 



LoPHUROMYS aquilus laticeps Thomas & Wroughton 

Plate 5, fig. 2. 

Lophnromys laticeps Thomas & Wroughton, 1907, Ann. Mag. Nat. Hist. (7), 
19, p. 383: Lake Kivu, Belgian Congo. 

1 cf 1 9 (M. C. Z. 40798-9) Mabira Forest, U. 9 & 17. xi. 38. 

1 cT juv. (M. C. Z. 39318) Bundibugyo, U. 20. xii. 38. 

4 d' 3 9 (M. C. Z. 39220, 39312-7) Mihunga, U. 10-19. i. 39. 

2 c? (M. C. Z. 39221-2) Nyakabande, U. 27. i. 39. 
1 cf 3 9 (M. C. Z. 39250-3) Kisenyi, B. R. 10. ii. 39. 
7 cf 6 9 (M. C. Z. 39254-66) Idjwi Id., B. C. 18. ii-3. iii. 39. 

Native names. Adulo (Luamba); kihukuzi (Lutoro); Jcisuhura (Lu- 
konjo); ichumba (Lulega). 

Dismission. Individual variation is so great that races of aquilus 
are somewhat questionable, young ones are brighter reddish below 
than most adults so that the appearance of a series depends to some 
extent on the average age. The Kisenyi and Idjwi series may be con- 
sidered topotypic of laticeps, to which Hatt assigns his Kisenyi ma- 
terial. Our four specimens from Kisenyi, on the north shore of the 
lake, are a trifle more ochraceous tawny on the under surface than the 
Idjwi series in which the tips of the belly hairs are tawny olive, and so 
short that the pale-gray bases show through, giving a gray-speckled 
appearance; a few individuals in the Idjwi series, however, are indis- 
tinguishable from those of Kisenyi. Apparently, to judge by his map, 
Hatt would apply the name rubecula (discussed above) to the rats of 
Ruwenzori; it appears impossible, however, to separate our seven 
Mihunga, Ruwenzori, from the Kivu material. The Ruwenzori 
series is very dark above, with minute speckling, and buffy gray or 
bright orange-buff below. Thomas & Wroughton (1910, p. 512) 
referred their Ruwenzori material to aquilus, their single Kivu speci- 
men to laticeps. 

All are more ochraceous on flanks and shoulders, less blackish on the 
back, than specimens from Kisiki, Belgian Congo, representing L. a. 
rita, towards which they are perhaps a transitional form. 



ALLEN AND LOVERIDGE: AFRICAN MAMMALS 205 

Measurements, cf. 141. 82. 20. 16 mm., 9- 138. 81. 19. 16 mm. 
Both from Idjwi. 

Breeding. In late February, a female held three fetuses circa 43 mm. 
from snout to anus. On March 3, I found a nest constructed of fine 
grass and large dry leaves, in a spacious cavity formed by the decaying 
roots of a large tree, which was situated in a small patch of forest. 
The nest held a 120 mm. female and a younger 9 • 100. 53. 23. 13 
mm., which was photographed, vide pi. 5, fig. 2. 

Diet. Taken with meat bait. 

Parasites. The male' from Mabira, like Arvicanthis* in the same 
locality, had numerous scars upon its back as if it had been parasitized 
by Tumbo fly. Five small ants on a trapped male at Mihunga, behaved 
Hke fleas in its fur. 

Enemies. On February 19, two new-born young in the stomach of a 
cobra {Naja melajwleuca) and four in that of a mamba {Dendroaspis 
j. haimosae), two adults in another mamba, one in a viper {Bitis 
nasicornis) at Mabira. 



LoPHUROMYS woosNAMi woosNAMi Thomas 

Lo-phuromys woosnami Thomas, 1906, Ann. Mag. Nat. Hist. (7), 18, p. 146: 
Mubuku Valley, Ruwenzori Mountains, Uganda, 6300 ft. 

c? 9 (M. C. Z. 39219, 39311) Mubuku Valley, U. 29. xii. 38-4. i. 39. 
9 (M. C. Z. 39218) Mihunga, Ruwenzori Mtns., U. 19. i. 39. 

Discussion. In addition to the topotypes listed above, there is a 
third preserved in alcohol. 

Measurements. &. 132. 113. 23. 20 mm., 9- 120. 97. 26. 13 mm., 
both from Mubuku, latter in alcohol. 

Habitat. The alcoholic specimen was trapped at 8 p.m. with cheese 
bait, the trap being set opposite its hole which was far beneath an 
overhanging rock in the heart of the forest. Footprints in the dust 
beneath the rock, attracted attention to the fact that the burrow was 
occupied. 

AcoMYS wiLSONi wiLSONi Thomas 

Acomys unlsotn Thomas, 1892, Ann. Mag. Nat. Hist. (6), 10, p. 22: Mombasa, 
Kenya Colony. 

9 (M. C. Z. 38954) Siga Caves, near Tanga, T. T. 14. vi. 39. 



206 bulletin: museum of comparative zoology 

Discussion. This specimen is darker on the back and brighter on 
the sides than a series of topotypes from Mombasa, but may be in- 
cluded within the range of individual variation. 

Measurements. 9 . 90. 52. 10. 12 mm. 

Habitat. Caught beneath palm fronds; its ears already in this ragged 
state. 

AcoMYS ? albigena Heuglin 

Acomys albigena Heuglin, 1877, Reise in Nordost-Afrika, 2, p. 69: Bogos 
country, Eritrea. 

9 (M. C. Z. 39058) Magrotto Mtn., T. T. 16. vii. 39. 

Native name. Kiberakandesi (Kisambara). 

Discussion. The single specimen, lacking a tail when trapped, is 
only tentatively referred to this species. It agrees closely in its dark 
blackish median area and bright orange-ochraceous sides with a skin 
from Dembea, Ethiopia, supposed by Dr. W. H. Osgood to represent 
Heuglin's species, but is widely different from A. ignitus, which, from 
geographical considerations, it might be expected to be. 

Measurements. 9 . 110. ?. 15. 15 mm. 

Diet. Stomach contents was largely insect remains, mostly un- 
identifiable, but Dr. P. J. Darlington, who kindly examined them, 
detects both medium and small-sized beetles. 



Dasymys bentleyae medius Thomas 

Dasymys medius Thomas, 1906, Ann. Mag. Nat. Hist. (7), 18, p. 143: Mubuku 
Valley, Ruwenzori Mountains, Uganda, 6000 ft. 

2 d' 2 9 (M. C. Z. 39224-5, 40832-3) Mihunga, U. 13-17. i. 39. 
1 9 (M. C. Z. 39149) Idjwi Id., L. Kivu, B. C. 3. iii. 39. 

Native name. Chumha (Lulega). 

Discussion. The Ruwenzori series is topotypic, and the Lake Kivu 
specimen is similar to them though with a slightly longer skull than 
any. We follow Hollister and Hatt in regarding medius as a race of 
bentleyae. 

Measurements. &. 130. 123. 28. 21 mm., 9. 155. 180. 30. 20 mm. 
Both from Mihunga, Ruwenzori, 6000 ft. 

Breeding. On March 3, a 155 mm. female was suckling two young, 
of which the cf measured 57. 28. 11.5 mm. (not kept). 



ALLEN AND LOVERIDGE: AFRICAN MAMMALS 207 

Enemies. A large adult recovered from the stomach of a mamba 
{Dendroaspis j. kaimosae) on Idjwi. 



Pelomys fallax concolor Heller 

Pelomys fallax concolor Heller, 1912, Smithsonian Misc. Coll., 59, No. 16, 
p. 13: Kduha, Lake Mutanda, Uganda. 

& (M. C. Z. 39146) Kisenyi, B. R. 11. ii. 39. 
d" 9 (M. C. Z. 39145, 40743) Idjwi Id., B. C. 22. ii. 39. 

Native name. Kinosa (Lulega). 

Discussion. These three specimens, one (M. C. Z. 40743) of which is 
young, agree in lacking the dorsal black stripe and the white beneath 
the forearms and on the belly, as described by Heller, but the sup- 
posedly narrower nasals, longer tooth row, and incisive foramina as 
compared with P. /. insignatus do not hold and may have been due to 
comparison with a specimen of Mylomys. In this genus, the validity 
of which is questionable, the incisive foramina extend backward 
between the anterior ends of the first molars, instead of stopping on a 
line with their anterior roots. 

Measurements. &. 182. 145. 30. 19 mm., 9. 105. 85. 22. 15 mm., 
both from Idjwi. 



Pelomys fallax insignatus Osgood 

Pelomys fallax insignatus Osgood, 1910, Ann. Mag. Nat. Hist. (8), 5, p. 276| 
Fort Hill, northern Nyasaland. 

1 9 (M. C. Z. 38842) Nchingidi, T. T. 17. v. 39. 

2 9 (M. C. Z. 38956-7) Magrotto Mtn., T. T. 17. vii. 39. 

Native name. Mende (Kisambara). 

Discussion. This race lacks the black dorsal stripe and has whitish- 
tipped hairs over the inguinal area and under surface of the forearms. 
In the youngest specimen (M. C. Z. 38957) there is a broad dark band 
in the mid-dorsal area where the yellow-ringed hairs have not yet come 
in. These specimens agree closely with one from Tukuyu, in the south- 
western part of Tanganyika Territory, secured by Loveridge in 1930, 
and now extend the recorded range to the coastal area in northeastern 
Tanganyika. 

Measvrcmerts. 9 . 158. 131. 28. 18 mm., from Magrotto Estate. 



208 bulletin: museum of comparative zoology 



Arvicanthis abyssinicus nubilans Wroughton 

Arvicanthus (sic) abyssinicus nubilans Wroughton, 1909, Ann. Mag. Nat. Hist. 
(8), 4, p. 539: Kisumu, Kenya Colony. 

3 cf 3 9 (M. C. Z. 40802-7) Mabira Forest, U. 12-19. xi. 38. 

2 d' 3 9 (M. C. Z. 39226-8, -30-1) Budongo Forest, U. 25. xi. 38. 

3 cf 5 9 (M. C. Z. 39291-8) Bundibugyo, U. 20. xii. 38. 

1 & (M. C. Z. 39229) Bugoye, U. 23. i. 39. 

1 & (M. C. Z. 39147) Nyakabande, U. 7. ii. 39. 

Native names. Mese (Luganda) ; Mhahu (Luamba) ; mheba (Lutoro) • 

Discussion. The series is uniform in color, with a faint buffy tinge 
over the shoulders, due to the pale-ochraceous subterminal bands on 
the hairs, deepening on the lower back to ochraceous. The dorsal 
black line is barely indicated, though traceable. 

Measurements. &. 175. 135. 32. 21 mm., 9- 160. 115.28.20 mm., 
both from Mabira. 

Eremics. One Mabira female had lost a hind leg but the skin was 
completely healed. 



Lemniscomys striatus massaicus (Pagenstecher) 

Mus {Lemniscomys) barbarus L. var. 7nassaicus Pagenstecher, 1885, Jahrb. 
Hamburg. Wiss. Anstalt, 2, p. 45: Lake Naivasha and Nguruman, Kenya 
Colony. 

2 cf (M. C. Z. 40800-1) Mabira Forest, U. 9. xi. 38. 

1 d^ 1 9 (M. C. Z. 39212,-302) Kibale Forest, U. 15. xii. 38. 

2 c? 2 9 (M. C. Z. 39303-6) Bundibugyo, U. 20. xii. 38. 

2 cf 1 9 (M. C. Z. 39211,-3,-301) Mihunga, U. 12-18. i. 39. 
2 young (M. C. Z. 39214-5) Bugoye, U. 23. i. 39. 

Native names. Mende (Luganda) ; besi (Luamba) ; nyaruveri (Lutoro) ; 
lusense (Lukonjo). 

Discussion. The Ruwenzori skins tend to be decidedly more gray- 
ish on nape and flanks, lacking the ochraceous tint of the others. The 
length of the hind foot, without claws, in the dried skins of all these 
specimens, is uniformly 25-26 mm., hence none is referable to L. m. 
macculus of Mokia,' southeast of Ruwenzori, in which the foot is 
21-23 mm. 

Measurements, d" . 130. 136. 26. 17 mm., 9. 127. 130. 25. ?. mm., 
both from Mihunga. 

1 Mohokya, twenty miles south of Bugoye. 



ALLEN AND LOVERIDGE: AFRICAN MAMMALS 209 

Breeding. On November 14, four nestlings (measuring 50. 35. 12. 
5 mm.) were found by men engaged in clearing undergrowth at Mabira. 

On December 15, a 125 mm. female held five small fetuses, at 
Kibale. 

On December 20, a 122 mm. female held five small fetuses, at 
Bundibugyo. 

On January 18, two blind nestlings were brought in at Mihunga. 

On January 23, two young (measuring 64. 45. 15. 7 mm.), their eyes 
still unopened, were brought in at Bugoye. 

On January 25, young were brought in, but not preserved, at 
Nyakabande. 

Parasites. Mites ( ) and maggots {Stasisia rodhaini) 

removed from myiasis-infected rats at Mabira. 

Otomys denti Thomas 

Otomys denti Thomas, 1906, Ann. Mag. Nat. Hist. (7), 18, p. 141 : Mubuku 
Valley, Ruwenzori Mountains, Uganda, 6000 ft. 

2 cf (M. C. Z. 39204-5) Mihunga, U. 4 & 18. i. 39. 

Native names. Kitwamusanzi (Lukonjo); kihukuzi (Lutoro). 

Discussion. These topotypes, having been taken at the Ruwenzori 
Expedition's (1905) camp, agree perfectly with Thomas's description 
in having the outer section of the lower incisors white, in having five 
laminae to the last upper molar, in their all-black tails, and under side 
darker, and with much less ochraceous speckling, than in 0. kempi. 

Measurements, cf. 160. 95. 27. 21 mm. 

Otomys kempi Dollman 

Otomys kempi Dollman, 1915, Ann. Mag. Nat. Hist. (8), 15,p. 152: Mt. Mikeno, 
Belgian Ruanda, 6000 feet. 

2 & (M. C. Z. 39299-300) Mabira Forest, U. 12 & 15. xi. 38. 
1 cf (M. C. Z. 39206) Kibale Forest, Toro, U. 13. xii. 38. 
1 9 (M. C. Z. 39148) Idjwi Id., L. Kivu, B. C. 3. iii. 39. 

Discussio7i. These specimens add slightly to the recorded distribu- 
tion, though Dollman has reported kempi from Buhamba, near Lake 
Kivu. The species is characterized by the possession of one groove on 
the lower incisor and six laminae to the last upper molar. Contrary to 
Dollman's account, however, the portion of the lower incisor external 
to the groove is not always white but is yellow in all four of the above 



210 bulletin: museum of comparative zoology 

specimens, and the tails are distinctly pale beneath. The lower surface 
has the bases of the hairs less blackish and with longer, more abundant 
buffy tips than 0. denti. 

Measurements, d^. 165. 75. 26. 20 mm., from Mabira. 



BATHYERGIDAE 

Cryptomys hottentotus whytei (Thomas) 

Georychus whytei Thomas, 1897, Froc. Zool. Soc. London, p. 432: Karonga, 
Lake Nyasa, Nyasaland. 

d' (M. C. Z. 38955) Ujiji, T. T. 13. iii. 39. 

Discussion. The single specimen is in very thin and much-worn 
pelage so that the true color characters are indeterminable, but shows 
four small white areas on the forehead. It agrees in its small foot and 
general cranial characters with specimens from southwestern Tangan- 
yika, which we previously (1933, p. 124) referred to C. h. ivhytei. The 
premaxillaries slightly exceed the nasals in posterior extension, but 
do not approximate each other or close behind them as they do in the 
occlusus of Kigogo. In May 1930, on an earlier visit to Lake Tangan- 
yika, Loveridge also obtained a solitary example of this species at 
Ujiji, which must be about the northern limit of its range. 

Measurements, cf. 145. 18. 19. 2 mm. 

HYSTRICIDAE 

Hystrix africaeaustralis ?subsp. 
c^ (M. C. Z. 39407) Kitaya, T. T. 2. iv. 39. 

Native names. Ndinu (Kiyao); iiungu (Kimakonde). 

Discussion. This is a young specimen with only the two anterior 
cheek teeth in place on each side, so that it offers no characters that 
would ensure its reference to any one of the four races named from 
Tanganyika Territory. In spite of its youth, the length of the frontals 
is more than half that of the nasals and twice the distance from the 
posterior edge of the frontals to the point of the occiput, thus agreeing 
with the adult H. africaeaustralis. 

Measurements, cf juv. 320. 60. 60. 34 mm. 

Enemies. This little creature with both fore and hind foot missing 
on the right side, though the stumps are entirely healed, was found in a 
maize shamba, and brought to me alive. 



ALLEN AND LOVERIDGE: AFRICAN MAMMALS 211 

LEPORIDAE 

Lepus capensis ?abbotti Hollister 

Lepus capensis abhotti Hollister, 1918, Proc. Biol. Soc. Wasliington, 31, p. 35: 
Plains east of Mt. Ivilimanjaro, Tanganyika Territory. 

cf 9 (M. C. Z. 38878-9) Mikindani, T. T. 15. iv. 39. 

Frequently seen at night on the roads near Mbanja, near Lindi, T.T. 

Native names. Chungula (Kimwera); usungula (Kimakonde at Kit- 
aya); ■uyungula (Kimakonde at Mbanja). 

Discussion. These two leverets are so young that they are referred 
to the race abhotti on geographical grounds only. As is usual in young 
hares, both have a prominent white blaze on the forehead. 

Measurements. cT and 9 . 190. 25. 56. 47 mm. Though external 
dimensions of both were alike, the skull of the male was surprisingly 
larger than that of the female. 

SLTDAE 

Hylochoerus meinertzhageni meinertzhageni Thomas 

Hylochoerus meinertzhageni Thomas, 1904, Nature (London), 70, p. 577: 
Kakamega Forest, near Kaimosi, Kenya Colony (see Loveridge, A., in 
Allen, G. M., and Lawrence, B., 1936, Bull. Mus. Comp. Zool., 79, p. 109). 

2 skulls (M. C. Z. 39428-9) Nyinabitaba, U. 19. i. 39. 

Habitat. Quite by accident, as we were packing up to leave the 
Ruwenzori Mountains, I learned that the Bakonjo were regularly 
killing giant forest hogs at Nyinabitaba, higher up the mountain above 
the Mubuku Valley. The skulls were prepared from heads which the 
hunters were carrying past our camp. 

BOVIDAE 

Cephalophus caerulus aequatorialis Matschie 

Cephalolophus (sic) aequatorialis Matschie, 1892, Sitzb. Ges. Naturf. Freunde 
Berlin, p. 112: Chagwe, Uganda. 

9 (M. C. Z. 39405) Mabira Forest, Chagwe, U. 16. i. 38. 
9 (M. C. Z. 39406) Kibale Forest, Toro, U. 15. xii. 38. 

Native names. Ntalaganiya (Liiganda) ; nencli (Lutoro and Lu- 
konjo). 



212 bulletin: museum of comparative zoology 

Discussion. The topotype from Mabira has a small rudiment of a 
horn core at the posterior margin of each frontal. It still retains the 
milk premolars and the last permanent molar is nearly in place. 

Measurements. 9 • 620. 100. 160. 52 mm., from Mubango, Mabira 
Forest. 

Breeding. Neither of these blue forest duikers was pregnant. 

Sylvicapra GR.IMMIA ?R00SEVELTi Heller 

Sylvicapra grimmia roosevelti Heller, 1912, Smithsonian Misc. Coll., 60, No. 8, 
p. 9: Ehino Camp (former Lado Enclave), Uganda. 

cf juv. (M. C. Z. 39424) Bundibugyo, U. 20. xii. 38. 

Native name. Abudi (Luamba). 

Discussion. This is a very young duiker with milk dentition, con- 
sequently its subspecific identification with roosevelti is solely on 
geographical probabilities and remains uncertain. 

Measurements, d*. 350. 55. 142. 46 mm. 



ALLEN AND LOVERIDGE: AFRICAN MAMMALS 213 



BIBLIOGRAPHY 

Allen, Glover M. 

1939. "A Checklist of African Mammals." Bull. Mus. Comp. Zool., 
~ 83, pp. 1-763. 

Allen, Glover M., and Loveridge, A. 

1927. "Mammals from the Uluguru and Usambara Mountains, Tan- 
ganyika Territory." Proc. Boston Soc. Nat. Hist., 38, pp. 413- 
44L 

1933. "Reports on the Scientific Results of an Expedition to the South- 
western Highlands of Tanganyika Territory. II. Mammals." 
Bull. Mus. Comp. Zool., 75, pp. 45-140, pi. i. 

Allen, Glover M., and Lawrence, B. 

1936. "Scientific Results of an Expedition to Rain Forest Regions in 
Eastern Africa. III. Mammals." Bull. Mus. Comp. Zool., 79, 
pp. 31-126, pis. i-v. 

Allen, J. A. 

1922. "Sciuridae, Anomaluridae, and Idiuridae collected by the American 
Museum Congo Expedition." Bull. Am. Mus. Nat. Hist., 47, 
pp. 39-71, pi. v. 
1924. "Carnivora collected by the American Museum Congo Expedi- 
tion." Bull. Am. Mus. Nat. Hist., 47, pp. 73-281, figs. 1-67, pis. 
vi-lxxviii, map. 

DoLLMAN, Guy 

1915. "On the African Shrews belonging to the Genus Crocidura. III." 
Ann. Mag. Nat. Hist. (8), 16, pp. 66-80. 

Hatt, R. T. 

1934. "The Pangolins and Aard-Varks collected by the American 
Museum Congo Expedition." Bull. Am. Mus. Nat. Hist., 66, 
pp. 643-672, figs. 1-2, pis. xxxii-.xxxix. 

1940. "Lagomorpha and Rodentia, other than Sciuridae, Anomaluridae, 
and Idiuridae, collected by the American Museum Congo Ex- 
pedition." Bull. Am. Mus. Nat. Hist., 76, pp. 457-604, figs. 1-39, 
pis. vii-xix, maps 1-2. 



214 bulletin: museum of comparative zoology 

HoLLiSTER, Ned 

1918. "East African Mammals in the United States National Museum. 

I. Insectivora, Chiroptera, and Carnivora." U. S. Nat. Mus. 
Bull., 99, pp. 1-194, figs. 1-3, pis. i-lv. 

1919. "East African Mammals in the United States National Museum. 

II. Rodentia, Lagomorpha, and Tubulidentata." U. S. Nat. Mus. 
Bull., 99, pp. 1-184, fig. 1, pis. i-xliv. 

Lawrence, Barbara 

1941. "Incisor tips of yoimg rodents." Zool. Ser. Field Mus. Nat. Hist., 
27, pp. 313-317, figs. 56, 57. (On p. 314, fines 3 and 8, for "total 
length" read head and body length.) 

ScHWARZ, Ernst 

1931. "On the African Long-tailed Lemurs or Galagos." Ann." Mag. 
Nat. Hi.«t. (10), 7, pp. 41-66. 



EXPLANATION OF PLATES 



PLATE 1 



Allen and Loveridgb — African Mammals. 



PLATE 1 
Map showing Principal Collecting Localities 

1938 

Landing at Mombasa (25.x), except for a stopover at Naivasha and Kin- 
angop (26-31. x), Loveridge proceeded by rail direct to Jinja (1-5. xi). Thence 
to Mabira Forest (5-21. xi), Budongo Forest (22.xi-7.xii), Kibale Forest 
(8-19.xii), Bundibugyo near Bwamba Forest (19-26.xii), Bugoye, foot of 
Ruwenzori Mountains (26-28.xii) and Mubuku Valley at 7000 ft. (29.xii-). 

1939 

On leaving Mubuku (-9.1) Loveridge descended down the valley to Mihunga, 
circa 6000 ft. (9-21. i), then back to Bugoye (21-24.i), Nyakabande (25-30.i), 
Mushongero (30.i-4.ii), returned to Nyakabande (4-8. ii); Kisenyi (8-13.ii), 
Goma (13-14. ii), Mamvu on Idjwi Lsland (14-16.ii), Upper Mulinga on Idjwi 
(16.ii-6.iii), Uvira (7-8.iii), Ujiji (9-16.iii), Dar es Salaam (18-19.iii), Mikin- 
dani (22-24.iii), Kitaya (24.iii-7.iv), Mikindani (7-24.iv), Mbanja (25.iv- 
6.v), Lake Rutamba (6-8.v), Nchingidi (9-21. v), Lindi (22.v-4.vi), Siga 
Caves (7-17.vi), Amboni Estate (17-27.vi), Magrotto Mountain (27.vi-21. 
vii), Tanga (21-23.vii), Kilindini (24-26.vii). 



BULL. MUS. COMP. ZOOL. 



Allen AND LovERiDGE. African Mammals. Plate 1. 



^t? 






niyaKahaniei -KabaU 



Uvira 



BELGIAN 
CONGO 




'^\Cii^ forest I ^l£ 




Siga Caves* X Q 



TAN G AN YH 
TER Rl TOR Y 






O 



^ 



,-< 



^ 



<^ 



<^ 



^l 








an2;bar 

lAR ES SALAAM 



^Ujindani 



Scale in Miles 





l_L- 



100 

_LJ 



200 

__l 



PLATE 2 



Allen and Loveridce — African Mammals. 



PLATE 2 

Fig. 1. Chimpanzee {Pan troglodytes schweinfurlhi) 

Apart from this young animal, offered for sale by its native owner at Bundi- 
bugyo, the only living chimpanzee seen during the course of the expedition 
was near Mihunga, on the eastern slopes of Ruwenzori. There, one afternoon, 
an old animal climbed a lone tree on a knoll and watched us intently as we 
hunted frogs in the swamp scarcely two hundred yards away. 

Fig. 2. Patas Monkey {Erythrocehus patas pyrrhonotus) 

This was a captive animal in the possession of Mr. J. L. Jarvis, on whose 
estate near Jinja it was photographed. No other Patas monkeys were seen 
during the present expedition, their range, in general, being further north. 
On a previous safari they were frequently seen in the thombush regions of 
Karamojo. 



BULL. MUS. COMP. ZOCL. 



Allen AND LovERiDdE. African Mammals. Plate 2. 



¥ 



KT-t^ 





PLATE 3 



Allen and Loveridge — African Mammals. 



PLATE 3 

Fig. 1. Galago (Galago demidovii thomasi) 

These small galagos make delightful pets. One, though of another species, 
captured at Nchingidi, was carried about for a month until it escaped into the 
forest at Amboni. There its kind were so abundant that just before dawn 
several pairs of glowing eyes might be seen in the beam of a single electric 
torch. 

Fig. 2. Tree Pangolevj {Phataginus tricuspis mabirae) 

The Mabira Forest would appear to be the most easterly point reached by 
the African tree pangolin. This eastern form differs in certain respects 
from its Cameroon congener and is therefore described as new. The above 
photograph is of the male type which, having been captured by a native in 
the nearby forest, was brought into camp at Mubango. 



BULL MUS. COMP. ZOOL. 



Allen AND LovERiDGE. African Mammals. Plate 3. 





PLATE 4 



Allen and Lovehidge — African Mammals. 



PLATE 4 

Fig. 1. Tree Pangolin using its prehensile tail 

Unlike the pangolin of the savanna, this species is associated with forest 
where it is said to climb trees in search of the arboreal termites on which it 
preys. Its powerful claws serve a useful purpose in digging out the termitaria, 
after which the elongate and viscous tongue is employed to take up the insects. 

Fig. 2. Tree P.\ngolin in defensive attittjbe 

Despite a certain similarity in defensive technique and general appearance, 
the African pangolins are not near relatives of the South American armadillos. 
When teased they are apt to uncoil and recoil with such startling suddenness 
that a bystander may receive quite a blow from the shielded tail. 



BULL. MUS. COMP. ZOOL. 



Allen AND LovERiDGE. Atrican Mammals. Plate 4. 















PLATE 6 



Allen and Lovehidge — African Mammals. 



PLATE 5 
Fig. 1. Banana Mice {Dendromus messorius ruddi) 

These attractive little creatures, their eyes already open, were found in their 
nest at Bugoye on January 23. As we were about to leave, we took them on to 
Nyakabande where we released them a week later. 

Fig. 2. Harsh-furred Mouse {Lophuromys aquilus laticeps) 

Together with its mother, this well-grown youngster was found in a nest 
composed of fine grass and dry leaves deep in a cavity among the roots of a 
forest tree on Idjwi Island, Lake Ivivu. 

Fig. 3. Dormouse (Claviglis murinus soleatus) 

In many parts of the country these gentle little gray dormice are not so 
common as formerly, according to the natives, who attribute their diminution 
to the spread of the omnivorous roof rat {Ratttis r. kijahius). 



BULL. MUS. COMP. ZOOL. 



Allen AND LovERiDGE. African Mammals. Plate 5. 





•V a* 




Bulletin of the Museum of Comparative Zoology 

AT HARVARD COLLEGE 

Vol. LXXXIX, No. 5 



SCIENTIFIC RESULTS OF A FOURTH EXPEDITION 
TO FORESTED AREAS IN EAST & CENTRAL AFRICA 

II 

BIRDS 



By James Lee Peters and Arthur Loveridge 



With Three Plates 



CAMBRIDGE, MASS., U. S. A. 

PRINTED FOR THE MUSEUM 

February, 1942 



No. 5. — Scientific Results of a Fourth 
Expedition to Forested Areas in East and Central Africa 

II 

Birds 
By James Lee Peters and Arthur Loveridge 

CONTENTS Page 

Introduction 217 

Index to Families collected 219 

Systematic discussion 221 

INTRODUCTION 

The collection on which the following report is based, was made by 
the junior author while investigating the herpetological fauna of cer- 
tain forested regions in East and Central Africa. The enquiry was 
carried out on behalf of the Museum of Comparative Zoology with a 
fellowship granted by the John Simon Guggenheim Memorial Founda- 
tion of New York. 

The period of collecting birds was from November 8, 1938, to July 
13, 1939, during which time 809 skins representing 246 species or races 
of birds were secured. Though 63 of these had never been taken before 
by the junior author, as the localities visited were in new faunal areas, 
only 10 were new to the collections of the Museum of Comparative 
Zoology, which are rapidly nearing completion for this general region. 

Of the additions, only one, an Apalis, of which eight examples were 
obtained on Idjwi Island, Lake Kivu, is described as new (p. 252). 
Besides discussions on variation and synonymy, field notes are included 
on diet, breeding, etc. As on previous occasions, considerable trouble 
was taken to secure the names applied by the various tribes, in ivhose 
district the species in question, ivas secured, as such names are frequently 
not being learned by the rising generation of natives who are more 
concerned with the impact of civilization. 

A synopsis of the itinerary is given in the caption accompanying 
plate 1 — a map showing principal collecting localities. Altitudes and 
detailed information regarding the various camps will be furnished in 



218 bulletin: museum of comparative zoology 

the final report of this series dealing with the general conclusions 
arrived at regarding distribution. 

For permission to use the blocks of plates 2 and 3, we are indebted 
to the Editor of the Scientific Monthly, in which journal (June and 
July, 1940) they appeared as illustrations to a popular account of the 
safari. 

We take this opportunity of thanking Dr. J. P. Chapin for pointing 
out that the Apalis was new, for the loan of comparative material, 
and for giving us the benefit of his valued opinion on half-a-dozen 
doubtful forms. To Mr. R. E. Moreau we are indebted for scanning 
the manuscript and making helpful suggestions based on his unrivalled 
knowledge of the birds of eastern Tanganyika. Where differences in 
rendering of bird names occurs as between his recorded in the Usam- 
bara Mountains and those obtained on Magrotto Mts. they are to be 
attributed to local dialect, those obtained by Mr. Moreau being the 
purer. We are also grateful to our colleague Dr. Joseph Bequaert for 
identifying ectoparasites, and to Drs. B. Schwarz and J. T. Lucker, 
of the United States Department of Agriculture, for similar courtesies 
regarding entoparasites. 

A selection of duplicates of such species, including a pair of the new 
Apalis, as were collected in the Belgian Congo, will be sent to the 
Congo Museum, Tervueren, in appreciation of the action of His Ex- 
cellency the Governor of the Congo Beige in granting permission to 
collect on Idjwi Island. 



PETERS AND LOVERIDGE: AFRICAN BIRDS 219 



INDEX TO FAMILIES 

Page 

ACCIPITRIDAE 223 

ALCEDINIDAE 236 

ANATIDAE 222 

ARDEIDAE 221 

BUCEROTIDAE 238 

CAMPEPHAGIDAE 244 

CAPITONIDAE 241 

CAPRIMULGIDAE 235 

CHARADRIIDAE 230 

CICONIIDAE 222 

COLIIDAE 236 

COLUMBIDAE 231 

CORACIIDAE 239 

CUCULIDAE 233 

DICRURIDAE 245 

FALCONIDAE 227 

FRINGILLIDAE 275 

GRUIDAE 229 

HIRUNDINIDAE 243 

INDICATORIDAE " 242 

LANIIDAE * . . , 261 

MEROPIDAE 237 

MOTACILLIDAE 259 

MUSCICAPIDAE 256 

MUSOPHAGIDAE 232 

NECTARINIIDAE 264 

NUMIDIDAE 229 

ORIOLIDAE 245 



220 bulletin: museum of comparative zoology 

Page 

PARIDAE 246 

PHALACROCORACIDAE 221 

PHASIANIDAE 228 

PICIDAE 242 

PHOENICULIDAE 240 

PLEGADIDAE 222 

PLOCEIDAE 267 

PSITTACIDAE 232 

PYCNONOTIDAE 247 

RALLIDAE 229 

SCOPIDAE 222 

STRIGIDAE 235 

STURNIDAE 263 

SYLVIIDAE 251 

TIMALIIDAE 246 

TRERONIDAE 230 

TURDIDAE 250 

TYTONIDAE 234 

UPUPIDAE 240 

ZOSTEROPIDAE "... 267 



PETERS AND LOVERIDGE : AFRICAN BIRDS 221 

Systematic Discussion 

PHALACROCORACIDAE 

Phalacrocorax carbo lugubris Riippell 
Phalacrocorax lugubris Riippell, 1845, Syst. Uebers., p. 134, pi. 50: Ethiopia. 
d' 9 (M. C. Z. 270601-2) Mushongero, U. 1. ii. 39. 

Both immature, the wings being 325 and 322 mm. respectively; 
foreneck brownish, while beneath the cf is black and the 9 mottled. 

Native name. Masovii (Lugezi). 

Breeding. These birds formed part of a huge colony on an island 
in Lake Mutanda, but inspection of the nests (pi. ii, fig. 2) appeared 
to indicate that the young had recently fledged. 

Diet. Catfish in stomach. 

HALiiiTOR AFRiCANUS AFRICANUS (Gmelin) 
Pelecanus africanus Gmelin, 1789, Syst. Nat., 1, pt. 2, p. 577: Africa. 
d^ (M. C. Z. 270603) Mushongero, U. 2 ii. 39. 
Native name. Lugano (Lugezi). 



ARDEIDAE 

Egretta garzetta dimorpha Hartert 

Egretta dimorpha Hartert, 1914, Bull. Brit. Orn. Club, 35, p. 14: southeast 
coast of Madagascar. 

cf 9 (M. C. Z. 270604-5) Siga Caves, T. T. 15. vi. 39. 

The d^ is in worn post-nuptial gray plumage with a single white 
feather in the left wing; wing 275 mm., bill 89 mm., tarsus 103 mm. 
The 9 is immature, the legs being black, the feet sea green. As the 
Museum of Comparative Zoology has a breeding pair of the white 
phase with nuptial plumes, shot at Dar es Salaam on 27.vi.18, the 
possibility of these birds from Siga, near Tanga, being wanderers from 
Madagascar, should not be overlooked. 

Diet. Stomach of cf held the following: 2 tadpoles of Rana houlen- 
geri, 2 young Rana m. mascareniensis, 2 young tailed Hemisus m. 
marmoratum, and 10 crabs; that of the female held indeterminate 
matter. 



222 bulletin: museum of comparative zoology 

SCOPIDAE 

Scopus umbretta bannermani Grant 

Scopus umbretta bannermani C. Grant, 1914, Bull. Brit. Orn. Club, 35, p. 27: 
Mt. Leganisho, Kenya Colony. 

? (M. C. Z. 270606) Mikindani, T. T. 18. iv. 39. 

Wing 305 mm. 

Native navie. Chengo (Kimakonde). 

Diet. A beetle (Dytiscus), many rat-tailed maggots (Eristalis), a 
grasshopper and numerous caterpillars. 

CICONIIDAE 

Dissoura episcopus microscelis (Gray) 

Ciconia microscelis G. R. Gray, 1848, Gen. Birds, 3, p. 561, pi. cli: No locality. 
South Africa (restricted Mackworth Praed). 

cf (M. C. Z. 270607) Mbanja, T. T. 2. v. 39. 

One of three woolly-necked storks feeding in the sea during forenoon. 
Native name. Korongo (Kimakonde) . 

PLEGADIDAE 

Hagedashia hagedash ? ERLANGERi Neumann 

Hagedashia hagedash erlangeri Neumann, 1909, Ornis, 13, p. 193: Dogge, South 
Somaliland. 

Breeding. On January 30, 1939, at Mushongero, Lake Mutanda, two 
hard-set eggs were brought in by a native, who pointed out the bird to 
me. The eggs are somewhat richer in coloring than an authenticated 
clutch obtained at Shanwa and the possibility of their belonging to 
the race nilotica should not be overlooked. 

ANATIDAE 

Anas undulata undulata Dubois 

Anas undulata Dubois, 1837, Orn. Gall., 1, p. 119, pi. Ixxvii: Cape of Good 
Hope. 

9 (M. C. Z. 270608) Nyakabande, U. 28. i. 39. 
d^, 3 9 9 (M. C. Z. 270609-12) Mushongero, U. 31. xii. 38 & 1. ii. 39 

Native name. Matanga (Lugezi). 



PETERS AND LOVERIDGE: AFRICAN BIRDS 223 

Nyroca erythrophthalma (Wied) 

Anas erythrophthalma Wied, 1832, Beitr. Nat. Bras., 4, p. 929: Lagoa do Bra50, 
near Villa de Belmonte, Bahia, Brazil. 

2 cf c^, 6 9 9 (M. C. Z. 270613-20) Mushongero, U. 1. ii. 39. 

M. Delacour (1937, Bull. B. O. C, 57, p. 157) has proposed 
Phoeoaythia as a substitute name for his genus Phoeonetta (pre- 
occupied) for this duck, formerly included in Nyroca. At the present 
time we hesitate to discuss as to whether the genus should be recog- 
nized. 

Native name. Mahinda (Lugezi). 

Dendrocygna viduata (Linnaeus) 
Anas viduata Linnaeus, 1766, Syst. Nat. ed. 12, 1, p. 205: Cartagena, Colombia. 
cf (M. C. Z. 270621) Kitaya, T. T. 28. iii. 39. 

Sarkidiornis melanota (Pennant) 
Anser melanotus Pennant, 1769, Ind. Zool., p. 12, pi., xi: Ceylon. 
9 9 (M. C. Z. 270622-3) Mushongero, U. 2. ii. 39. 
Native name. Mifewe (Lugezi). 

ACCIPITRIDAE 

Machaerhamphus alcinus anderssoni (Gurney) 

Stringonyx anderssoni Gurney, 1866, Proc. Zool. Soc. London for 1865, p. 618: 
Otjimbingue, Damaraland, Southwest Africa. 

d" (M. C. Z. 270624) Siga Caves, T. T. 12. vl. 39. 

Diet. Stomach held a bat (Coleura afra), the species being abundant 
in the caves near which the hawk was shot in the gloaming. 

MiLVUs MIGRANS MIGRANS (Boddaert) 

Falco migrans Boddaert, 1783, Tabl. PI. Enl., p. 28, no. 472: No locality. 
France (Hartert). 

Migration. On November 1, 1938, at Tororo, Uganda, large num- 
bers of these kites filled the trees both east and west of the station, 
about thirty birds occupied one tree near where the train stopped. 
On February 7, 1939, at Nyakabande, Uganda, a large flight arrived 
from the southeast at sunset to settle in nearby trees. 



224 bulletin: museum of comparative zoology 

AcciPiTER TACHiRO SPARSiMFASciATUs (Reichenow) 

Astur sparsimfasciatus Reichenow, 1895, Orn. Monatsb., 3, p. 97: Zanzibar. 

9 (M. C. Z. 270625) Alikindani, T. T. 19. iv. 39. 
9 (M. C. Z. 270626) Mbanja, T. T. 28. iv. 39. 

The latter trapped by a native, who has removed some of the flight 
feathers. 

Native name. Chiuhinda (Kimakonde). 
Breeding. Ovules small in the Mikindani bird. 
Diet. A moderate sized bird in stomach of same. 



AcCIPITER RADIUS POLYZONOIDES Smith 

Accipiter polyzonoides A. Smith, 1838, 111. Zool. S. Africa, Aves, pi. 11: "N. 
of 26° S. lat.," probably near Mafeking. 

& (M. C. Z. 270627) Mikindani, T. T. 11. iv. 39. 

Native name. Narumanga (Kimakonde). 

Diet. Stomach held three diurnal geckos {Lygodactylus p. pictura- 
tus). 

Accipiter minullus tropicalis Reichenow 

Accipiter minullus tropicalis Reichenow, 1898, Journ. f. Orn., p. 139: Eaist 
Africa. 

d" (M. C. Z. 270628) Mikindani, T. T. 14. iv. 39. 

Despite remarks by Friedmann (1930, Bull. 153, U. S. Nat. Mus., p. 
64) to the contrary, intermedins and tropicalis appear to us to be recog- 
nizable races. 

Native name. Naluvi (Kimakonde). 

Breeding. Testes small. 

Diet. Stomach held a gecko {Lygodactylus p. picturatus) and 
grasshopper. 

Melierax gabar (Daudin) 

Falco gabar Daudin, 1800, Traite, 2, p. 87: interior of S. Africa {ex Levaillant). 

9.(M. C. Z. 270629) Kitaya, T. T. 3. iv.39. 

Native name. Nancheka (Kimakonde). 

This bird, which has been examined by Dr. Chapin also, is in im- 
mature plumage. 



PETERS AND LOV BRIDGE: AFRICAN BIRDS 225 

BuTEO RUFOFUSCUS AUGUR (Riippell) 

Falco (Buteo) augur Riippell, 1836, Neue Wirbelth., Vogel, p. 38, pi. xvi: 
Ethiopia. 

cf (M. C. Z. 270630) Idjwi Id., B. C. 26. ii. 39. 

The melanistic form, shot on the mountain dominating the island. 
Breeding. On January 30, 1939, at Mushongero, Uganda, two 
sHghtly incubated eggs were brought in by a native. 
Diet. Stomach held rodent fur. 

BuTEO OREOPHILUS Hartert & Neumann 

Buteo oreophilus Hartert & Neumann, 1914, Orn. Monatsb., 22, p. 31: Korit- 
scha, s. Ethiopia. 

9 (M. C. Z. 270631) Mubuku ^'alley, U. 4. i. 39. 
9 (M. C. Z. 270632) Mihunga Eidge, U. 18. i. 39. 

Native name. Ibebe (Lukonjo and Lutoro). 
Diet. Stomach held mouse. 
Parasites. Lice preserved on 18.i.39. 

Kaupifalco monogrammicus (Temminck) 

Falco monogrammicus Temminck, 1824, PL Col, livr. 53. pi. cccxiv: Senegal. 

c^ 9 9 (M. C. Z. 270633-5) Kitaya, T. T. 28-30. iii. 39. 
cf (M. C. Z. 270636) Mikindani, T. T. 17. iv. 39. 
c^ (M. C. Z. 270637) Mbanja, T. T. 28. iv. 39. 
(f (M. C. Z. 270638) Amboni Estate, T. T. 14. vi. 39. 

All birds in the series are moulting, so wing measurements mis- 
leading. 

Native names. Chitotola (Kiyao); nalumanga (Kimakonde). 

Breeding. On March 28, 1939, both 9 9 held small ovules. 

Diet. Stomachs held frog, cricket {Brachytry petes membranaceus), 
and grasshopper at Kitaya; praying mantis at Mikindani; lizard {Gerr- 
hosaurus n. nigrolineatus), skink {Mabiiya v. varia), and grasshopper 
at Mbanja; lizard (G. n. nigrolineatus), blind snake (Typhlops u. 
unitaeniatus) , and grasshopper at Amboni. 

Lophaetus occipitalis (Daudin) 

Falco occipitalis Daudin, 1800, Traitc, 2, p. 40: Anteniquoi country, i.e. Knysna 
district, Cape Province. 

cf a" 9 (M. C. Z. 270639-41) Magrotto Mtn., T. T. 3 & 13. vii. 39. 



226 bulletin: museum of comparative zoology 

Native 7iame. Seuchungi (Kisambara). 

Diet. Stomach of one held rat fur, those of the others, shot at 
8 a.m. and 4 p.m. respectively, were empty. 

Stephanoaetus coronatus (Linnaeus) 

Falco coronatus Linnaeus, 1766, Syst. Nat. ed. 12, 1, p. 124: Guinea, i.e. West 
Africa. 

Diet. There were three crowned hawk-eagles frequenting the forest 
in the vicinity of my camp on Magrotto Mountain, and though they 
roosted in the trees just across the valley from my tent, they were 
always out of gunshot. One day as Mr. Clausen, manager of the near- 
by plantation, was approaching the oil-nut factory, one of these great 
birds dropped from the sky and seized the full-grown cat which lived 
at the factory. This animal throve on the numerous rats and was 
rather wild, so one might have supposed it would have been able to 
give a good account of itself; actually it was powerless in the formidable 
talons. The eagle carried it to a nearby tree, and Clausen, hoping to 
get the bird for me, dispatched a boy for his gun; before it arrived, 
however, the eagle, disturbed by some noisy natives, flew away. 

Aquila rapax rapax (Temminck) 

Falco rapax Temminck, 1828, PI. Col. livr. 76, pi. cccclv: South Africa. 
9 (M. C. Z. 270642) Nyakabande, U. 7. ii. 39. 
Breeding. Ova small. 

Necrosyrtes monachus pileatus (Burchell) 

Vultur pileatus Burchell, 1824, Travels, 2, p. 195: Country south of Orange 
River, i.e. Hopetown district, Cape Province. 

c3^ 9 (M. C. Z. 270643-4) Mikindani, T. T. 10. iv. 39. 

Native name. Dilcviho (Kimakonde). 
Breeding. Gonads small. 

Gypohierax angolensis (Gmelin) 

Falco angolensis Gmelin, 1788, Syst. Nat., 1, p. 252: Angola. 

cf cf (M. C. Z. 270645-6) Magrotto Mtn., T. T. 14. vii. 39. 

Native name. Nyumbun (Kisambara). 

Diet. Both stomachs full of palm nuts from the Magrotto Estate. 

Parasites. Hippoboscids {Lynchia dukei) in plumage. 



PETERS AND LOVERIDGE: AFRICAN BIRDS 227 

Gymnogenys typicus typicus (Smith) 

Poloboroides [sic] typicus A. Smith, 1830, S. Afr. Quart. Journ. (1), p. 107: 
Eastern Cape Province. 

c^ (M. C. Z. 270647) Kitaya, T. T. 28. iii. 39. 
9 (M. C. Z. 270648) Mikindani, T. T. 20. iv. 39. 
9 (M. C. Z. 270649) Mbanja, T. T. 28. iv. 39. 

Native name. Lipupilambugu (Kiyao); fiamakofigolo (Kimakonde 
at Kitaya); narvfi (Kimakonde at Mbanja). 

Breeding. On March 28, and April 28, testes and ovules small. 

Diet. Stomach of one held two nestling weavers which had been 
swallowed whole. 

CiRCAETUS CINEREUS Vieillot 

Circaetus cinereus Vieillot, 1818, N. Diet. d'Hist. Nat.. 23, p. 445: Senegal. 

9 (M. C. Z. 270650) Mbanja, T. T. 28. iv. 39. 

9 (M. C. Z. 270651) Amboni Estate, T. T. 24. vi. 39. 

Diet. Stomach of first held three snakes {Psammophis s. sudanen- 
sis); of the second, four snakes (P. s. sudanensis, P. s. sihilans, and 
Causus defiUippi). 

Parasities. Hippoboscids {Pseudolynchia canariensis) in plumage. 



FALCONIDAE 

Falco tinnunculus rufescens (Swainson) 

Falco rujuscens [sic] Swainson, 1837, Birds West Africa, 1, p. 109: Senegal 
(presumed) 

9 (M. C. Z. 270652) Mihunga Ridge, U. 12. i. 39. 

Owing to the difficulty of identifying single skins of this hawk, the 
senior author submitted it to Dr. J. P. Chapin, who writes: "This bird 
belongs to the dark resident race of the edges of the Ituri forest, which 
I have called rufescens and most other people carlo, including you in 
your Check List. It has always seemed to me that while females from 
N. E. Africa look very like females from N. E. Congo, the males in the 
homeland of carlo are much less heavily barred on the back than the 
bird which ranges from the Ituri west to Nigeria." 

Native name. Ibebe (Lutoro and Lukonjo). 

Diet. Stomach held a chameleon (C. b. ellioti). 



228 bulletin: museum of comparative zoology 

PHASIANIDAE 

Francolinus sephaena rovuma Gray 

Francolinus rovuma G. R. Gray, 1867, List Spec. Birds Brit. Mus., pt. 5, Gal- 
linae, p. 52: Rovuma River. 

cf (M. C. Z. 270653) Mikindani, T. T. 15. iv. 39. 
c^ (M. C. Z. 270654) Mbanja, T. T. 2. v. 39. 

9 

These almost topotypic birds possess conspicuous reddish brown 
shaft stripes over the entire under surface, in which respect they 
differ from F. s. granti of central Tanganyika Territory. 

Francolinus hildebrandti grotei Reichenow 

Francolinus grotei Reichenow, 1919, Journ. f. Orn., p. 334: Mikindani, Tan- 
ganyika Territory. 

c^ c^ (M. C. Z. 270655-6) Mikindani, T. T. 18. iv. 39. 

Being topotypic of grotei we use that name tentatively, though this 
race was recently synonymized with the Zomba bird, johnstoni, of 
which we have no material, by Grant and Mackworth-Praed (1935, 
Ibis, p. 670). 

Breeding. Attracted by its loud calling, I stalked one of these birds 
for a quarter of an hour, then shot it as it was fighting another cock. 

Francolinus squamatus zappeyi Mearns 

Francolinus schuetti zappeyi Mearns, 1911, Smiths. Misc. Coll., 56, no. 20, p. 4: 
East shore of Lake Victoria. 

9 (M. C. Z. 270657) Nyakabande, U. 28. i. 39. 

Breeding. This bird, together with her five eggs (preserved), was 
brought in by a native. 

Pternistis afer melanogaster Neumann 

Pternistes nudicollis melanogaster Neumann, 1898, Journ. f. Orn., 46, p. 299, 
pi. iii, fig. 1 : Tanga, Tanganyika Territory. 

d' (M. C. Z. 270658) Amboni Estate, T. T. 16. vi. 39. 

This bird is topotypic. The birds from Lumbo, Mozambique, 
referred to P. a. humboldtii by Friedmann (1937, Bull, Mus. Comp. 
Zool., 81, p. 60), have been examined by Grant and Mackworth-Praed 



PETERS AND LOVERIDGE: AFRICAN BIRDS 229 

(1936, Bull. Brit. Orn. Club, 57, pp. 43-4) and redetermined as 
melanogaster. 

Native name. Kivale (Kisambara). 

Diet. Stomach held mtama grain. 



NUMIDIDAE 

NUMIDA MELEAGRIS MITRATA Pallas 

Numida mitrata Pallas, 1767, Spic. Zool., 1, fasc. iv, p. 18, pi. iii: Madagascar. 
cf (M. C. Z. 270659) Mikindani, T. T. 18. iv. 39. 

In life, crest yellow; line down nape black; crown, back' of head, 
lores, and three spots below eye, red; behind eye and sides of neck pale 
blue; lappets blue, tipped with coral red; throat dark blue. 

GuTTERA EDOUARDi SETH-SMiTHi Neumann 

Guttera cristata seth-smithi Neumann, 1908. Bull. Brit. Orn. Club, 23, p. 13: 
Budongo Forest, Unyoro, Uganda. 

cf cf 9 (M. C. Z. 270660-2) Kibale Forest, U. 16-17. xii. 38. 



GRUIDAE 

Balearica pavonina gibbericeps Reichenow 

Balearica gibbericeps Reichenow, 1892, Journ. f. Orn., p. 126: Lake Jipe, near 
Kilimanjaro, Tanganyika Territory. 

juv. 9 (M. C. Z. 270663) Nyakabande, U. 27. i. 39. 

Breeding. The above (pi. ii, fig. 1) was one of several nestlings 
offered for sale at Nyakabande. On February 3, at Mushongero, Lake 
Mutanda, a clutch of hard-set eggs (one preserved) were brought in by 
a native. 

RALLIDAE 

LiMNOCORAX flavirostra (Swainson) 

Gallinula flavirostra Swainson, 1837, Birds West Afr.. 2, p. 244, pi. xxviii: 
Senegal. 

cf (M. C. Z. 270664) Kitaya, T. T. 31. iii. 39. 

Native names. Nachindi (Kiyao); nantikinya (Kimakonde). 



230 bulletin: museum of comparative zoology 

Sarothrura pulchra centralis Neumann 

Sarothrura pulchra centralis Neumann, 1908, Bull. Brit. Orn. Club, 21, p. 45: 
Mswa, west shore of Lake Albert, Belgian Congo. 

? cf (M. C. Z. 270665) Bundibugyo, U. 20. xii. 38. 

Native name. Mhulu (Luamba). 

CHARADRIIDAE 

Charadrius alexandrinus tenellus Hartlaub 

Charadrius tenellus Hartlaub, 1861, Fauna Madagascar, p. 72: Madagascar. 
9 (M. C. Z. 270666) Mikindani, T. T. 14. iv. 39. 

Native name. Limbapa (Kimakonde). 

Xiphidiopterus albiceps (Gould) 

Vanellus albiceps Gould, 1834, Proc. Zool. Soc. London, p. 45: Quorra River 
{i.e. Niger) or Fernando Po. 

cf (M. C. Z. 270667) Kitaya, T. T. 31. iii. 39. 

The white-headed plover is a western species, but has been recorded 
already from the Rufigi River, 200 miles north of the Rovuma where 
this was shot. 

TRERONIDAE 

Treron calva salvadorii (Dubois) 

Vinago salvadorii Dubois, 1897, Proc. Zool. Soc. London, p. 784: Eastern 
and central tropical Africa. West shores of Lake Tanganyika (restricted 
Hartert and Goodson). 

c? 9 (M. C. Z. 270668-9) Idjwi Id., B. C. 1. iii. 39. 

These birds are more or less intermediate between caha and salva- 
dorii, being yellower than the former yet not so yellow as the latter. 
Native name. Chokore (Lulega). 

Treron calva wakefieldii (Sharpe) 

Vinago wakefieldii Sharpe, 1874, Proc. Zool. Soc. London for 1873, p. 715, pi. 
iviii, fig. 2: Mombasa, Kenya Colony. 

cf (M. C. Z. 270670) Amboni Estate, T. T. 24. vi. 39. 

This seems best treated as a race of calva than as a separate species. 
Native name. Ninga (Kisambara). 



PETERS AND LOVERIDGE: AFRICAN BIRDS 231 

COLUMBIDx\E 

CoLUMBA ARQUATRix ARQUATRix Temminck 

Columba arquatrix Temminck, 1809, Pigeons, Colombes, p. 11, pi. v: Anteni- 
quoi country, i.e. Knysna district. Cape Province. 

9 (M. C. Z. 270671) Idjwi Id., B. C. 21. ii. 39. 

Native name. Shindamulala (Lulega). 

Streptopelia semitorquata semitorquata (Riippell) 

Columba semitorquata Riippell, 1837, Neue Wirbelth., Vogel, p. 66, pi. xxiii, 
fig. 2: Taranta Mountains, Ethiopia. 

& 9 (M. C. Z. 270672-3) Mihunga Ridge, U. 11. i. 39. 

cT 9 (M. C. Z. 270674-5) Nyakabande, U. 7. ii. 39. 

cf 9 (M. C. Z. 270676-7) Idjwi Id., B. C. 21. ii. & 1. iii. 39. 

9 (M. C. Z. 270678) Kitaya, T. T. 30. iii. 39. 

d'd' 9 (M. C. Z. 270679-81) Magrotto Mtn., T. T. 1. vii. 39. 

Native names. Dica (Lutoro); kalikuku (Lukonjo); njuva (Kiyao); 
lideya (Kimakonde); hua (Kisambara). 

Breeding. On February 7, ova enlarged \o 10 mm. diameter. 

Streptopelia capicola somalica (Erlanger) 

Turtur damarensis somalicus Erlanger, 1905, Journ. f. Orn., p. 127: Sarigo, 
southern Somaliland. 

9 (M. C. Z. 270682) Ivitaya, T. T. 28. iii. 39. 

This single specimen appears to agree better with somalica than 
with either tropica of nearby Songea, or ancej^s of Kilosa; in the absence 
of adequate series of all three, however, the determination need not 
be regarded as final. If correct, it involves a southward extension of 
range for somalica along the coast from Pangani River to the Rovuma 
River. 

Native names. Njuva jalusele (Kiyao); chimbelele (Kimakonde). 

Turtur chalcospilos chalcospilos (Wagler) 

Columba chalcospilos Wagler, 1827, Syst. Av., Columba, sp. 82: "Terra Caffro- 
rum," i.e. eastern Cape Province. 

4 cf cf (M. C. Z. 270683-6) Kitaya, T. T. 27. iii-3. iv. 39. 
c? (M. C. Z. 270687) Mikindani, T. T. 17. iv. 39. 
cT (M. C. Z. 270688) Mbanja, T. T. 28. iv. 39. 



232 bulletin: museum of compaeative zoology 

Native names. Kitukutuku (Kiyao) ; chindutu (Kimakonde at Kit- 
aya); lidea (Kimakonde at Mbanja). 

Breeding. On March 30, at Kitaya, 2 eggs measuring 20 x 16 mm. 
were collected. 

PSITTACIDAE 

PsiTTACUS ERiTHACUS ERiTHACUS Linnacus 
Psittacus erithacus Linnaeus, 1758, Syst. Nat. ed. 10, 1, p. 99: Guinea. 
d" 9 (M. C. Z. 270689-90) Idjwi Id., B. C. 1. iii. 39. 
Native name. Kamiku (Lulega). 

POICEPHALUS CRYPTOXANTHUS TANGANYIKAE Bowen 

Poicephalus fuscicapillus tanganyikae Bowen, 1930, Proc. Acad. Nat. Sci. 
Philadelphia, 82, p. 267: Ivilosa, Tanganyika Territory. 

1 (M. C. Z. 270691) Kitaya, T. T. 25. iii. 39. 
cf 9 (M. C. Z. 270692-3) Mikindani, T. T. 18. iv. 39. 
cf 9 (M. C. Z. 270694-5) Mbanja, T. T. 24. vi. 39. 

Wings measure: d^cf 147-149 mm., 9 9 139-141 mm. 

Grant and Mackworth-Praed (1938, Bull. B. O. C, 59, p. 27) sink 
tanganidae [sic] in the synonymy of cryptoxantkus after adding the 
wing length of a Tanganyika 9 to the range of the Mozambique race. 
There appears to be a definite average size difference, however, which 
would justify retention of tanganyikae. 

Native names. Ngicesi (Kiyao) ; mwhendi (Kimakonde at Kitaya) ; 
ungudi (Kimakonde at Mikindani). 

MUSOPHAGIDxlE 

Tauraco schuttii emini (Reichenow) 

Turacus em,ini Reichenow, 1893, Orn. Monatsb., p. 30: Bundeko, Semliki 
Valley, Belgian Congo. 

d^ 9 9 (M. C. Z. 270696-8) Budongo Forest, U. 2-3. xii. 38. 

MUSOPHAGA VIOLACEA ROSSAE Gould 

Musophaga rossae Gould, 1851, Proc. Zool. Soc. London, p. 93: Western coast 
of Africa, i.e. Loanda, Angola (fide Grant, 1915, Ibis, p. 413). 

3 c? cf 2 9 9 (M. C. Z. 270699-702) Idjwi Id., B. C. 18. ii-2. iii. 39. 
Native name. Nduku (Lulega). 



PETERS AND LOVERIDGE: AFRICAN BIRDS 233 

CORYTHAEOLA CRISTATA (VieiUot) 

Musophaga cristata Vieillot, 1816, Analyse, p. 68: Africa. 

3 cf cT (M. C. Z. 270703-5) Mabira Forest, U. 12-15. xi. 38. 
2 cf c^ (M. C. Z. 270706-7) Budongo Forest, U. 1. xii. 38. 
(f (M. C. Z. 270708) Kibale Forest, U. 13. xii. 38. 



CUCULIDAE 

CucuLUS CLAMosus CLAMOsus Latham 

Cundus damosus Latham, 1802, Genl. Syn., 2, Suppl., p. xxx: Cape of Good 
Hope, i.e. Cradock Division, Cape Province. 

3 cf cT 2 9 9 (M. C. Z. 270709-13) Kitaya, T. T. 25-28. iii. 39. 
9 (M. C. Z. 270714) Mbanja, T. T. 27. iv. 39. 

Native names. Nankolca (Kiyao); m.'pando (Kimakonde). 

Breeding. Of one cf a single testis only slightly enlarged. 

Diet. One stomach held three black-and-yellow, spiny caterpillars. 

Clamator cafer (Lichtenstein) 

Ciiculus cafer A. Lichtenstein, 1793, Cat. rer. rar. Hamb., p. 14: Kaffirland, 
i.e. eastern Cape Province. 

9 (M. C. Z. 270715) Kitaya, T. T. 27. iii. 39. 

Native name. Nanhoka (Kiyao, see above also) ; nanchuwi (Kima- 
konde). 

Chrysococcyx klaas klaas (Stephens) 

Cuculus klaas Stephens, 1815, in Shaw, GenL Zool., 9, p. 128: Platte River or 
Cape Province (fide Levaillant). 

d^ (M. C. Z. 270716) Mihunga ridge, U. 19. i. 39. 
d^ (M. C. Z. 270717) Kitaya, T. T. 25. iii. 39. 
cT (M. C. Z. 270718) Mikindani, T. T. 14. iv. 39. 

Native names. Nderemhi (Lukonjo); niwi (Kimakonde). 

Chrysococcyx caprius (Boddaert) 

Cuculus caprius Boddaert, 1783, Tabl. PI. Enlum, p. 40, no. 657: Cape of Good 
Hope. 

1 cf 3 9 9 (M. C. Z. 270719-22) Kitaya, T. T. 30. iii. 39. 



234 bulletin: museum of comparative zoology 

Ceuthmochares aereus intermedius Sharpe 

Ceuthmochares intermedins Sharpe, 1884, Journ. Linn. Soc. London, Zool., 17, 
p. 432: Semmio (i.e. Zemio), Niam-niam, Belgian Congo. 

d' (M. C. Z. 270723) Budongo Forest, U. 28. xi. 38. 
(f 9 (M. C. Z. 270724-5) Idjwi Id., B. C. 27. ii. 39. 

"Wings of c^ and 9 from Idjwi measure 118-120 mm., tails 219-222 
mm. 

Native name. Legashuku (Lulega). 

Centropus superciliosus loandae Grant 

Centropus superciliosus loandae C. Grant, 1915, Bull. Brit. Orn. Club, 35, p. 54: 
near Dalla Tando, Angola. 

9 (M. C. Z. 270726) Idjwi Id., B. C. 3. iii. 39. 
Native name. Chibilibili (Lulega). 

Centropus superciliosus burchellii Swainson 

Centropus burchellii Swainson, 1837, Anim. Menag., p. 321: South Africa, i.e. 
Cape Frovince (ex Burchell). 

cf juv., 9 (M. C. Z. 270727-8) Kitaya, T. T. 27. iii. 39. 
cf (M. C. Z. 270729) Mikindani, T. T. 18. iv. 39. 
c^ (M. C. Z. 270730) Mbanja, T. T. 2. v. 39. 
cf (M. C. Z. 270731) Nchingidi, T. T. 10. v. 39. 

T]ie last two birds come from localities ten miles north and fifty 
miles southwest of Lindi and are therefore almost topotypic of fascii- 
■pygialis Reichenow (described from Lindi; QuiHmane; Mozambique; 
etc.). They have also been compared with birds from Lumbo, Mo- 
zambique. We regard fasciipygialis as a synonym. 

Native names. Litipilipili (Kiyao); natipitcla (Kimakonde). 



TYTONIDAE 

Tyto alba affinis (Blyth) 

Strix affinis Blyth, 1862, Ibis, p. 388: Cape of Good Hope. 

c? 9 (M. C. Z. 270732-3) Nchingidi, T. T. 17. v. 39. 

Diet. Stomach of each held a bird. According to their native captor, 
both these owls were taken from a hollow tree. 



PETERS AND LOVERIDGE: AFRICAN BIRDS 235 

STRIGIDAE 

Otus leucotis granti (Kollibay) 

Pisorhina leucotis granti Kollibay, 1910, Orn. Monatsb., 18, p. 148: Southwest 
Africa. 

cf (M. C. Z. 270734) Mbanja, T. T. 2. v. 39. 

Diet. Stomach held a cricket and praying mantis. This owl flew 
into my bat net shortly after sunrise. 

Bubo africanus africanus (Temminck) 

Strix africana Temminck, 1823, PI. Col., livr. 9, pi. 50: Cape of Good Hope. 

9 (M. C. Z. 270735) Magrotto Mtn., T. T. 6. vii. 39. 

Native name. Kungioi (Kisambara). 

Diet. Stomach held an adult house rat {Rattus r. kijahim) entire! 

Ciccaba woodfordii nigricantior (Sharpe) 

Syrnium nigricantius Sharpe, 1897, Bull. Brit. Crn. Club, 6, p. xlvii : Mpwapwa, 
Tanganyika Territory. 

9 (M. C. Z. 270736) Amboni Estate, T. T. 21. vi. 39. 

Diet. Stomach held an acridian, grasshopper, and praying mantis. 
I shot this owl at night in a tree at the forest edge, thereby disturbing 
a troupe of colobus monkeys who made off with loud cries. 

CAPRIMULGIDAE 

Scotornis fossii mossambicus (Peters) 

Caprimulgvs mossambicus Peters, 1868, Journ. f. Cm., 16, p. 134: Inhambane, 
Mozambique. 

cf 9 (M. C. Z. 270737-8) Kitaya, T. T. 3. iv. 39. 
Native names. Nalumbapala (Kiyao); nalumbao (Kimakonde). 

Macrodipteryx longipennis (Shaw) 

Caprimulgus longipennis Shaw, 1796, Nat. Misc., 8, p. 265: Sierra Leone. 

9 9 (M. C. Z. 270739-40) Butiaba Escarpment, U. 29. xi. 38. 

A plume-bearing cf standard-winged nightjar was shot at the same 
time as the foregoing but was lost in long grass by the side of the road 
on which they were sitting at the summit of the escarpment. 



236 bulletin: museum of comparative zoology 

COLIIDAE 

CoLius STRIATUS MOMBASSicus Van Someren 

CoUus striatus mombassicus van Someren, 1919, Bull. Brit. Orn. Club, 40, p. 26: 
Changamwe, Kenya Colony. 

2 d'd' 2 9 9 (M. C. Z. 270741-44) Siga Caves, T. T. 10. vi. 39.) 

These colies have whiter cheeks than those of our Morogoro material 
representing affinis Shelley. 

CoLius STRIATUS KiwuENSis Reichenow 

Colius Mivuensis Reichenow, 1908, Orn. Monatsb., p. 191: Lake Kivu. 

d' d' (M. C. Z. 270745-6) Mihunga Ridge, U. 10. i. 39. 
d 9 (M. C. Z. 270747-8) Idjwi Id., B. C. 24. ii. 39. 

The latter are topotypes, the former within the range of this race 

as defined by Chapin (1939, Bull. Am. Mus. Nat. Hist., 75, p. 475). 

Native names. Msonono (Lutoro) ; visole (Lukonjo) ; kishile (Lulega). 

ALCEDINIDAE 

CORYTHORNIS CRISTATA CRISTATA (Pallas) 

Alcedo cristata Pallas, 1764, in Vroeg, Cat., Adumb., no. 55, pi. i: Cape of Good 
Hope. 

9 (M. C. Z. 270749) Mushongero, U. 3. ii. 39. 

Taken late at night as it was roosting on a papyrus just above the 
water; paddling up in a dugout, I picked it off by hand. 

IsPiDiNA PICTA PICTA (Boddacrt) 

Todus pictus Boddaert, 1783, Tabl. PI. Enlum., p. 49: "Juida," i.e. St. Louis, 
Senegal {fide Buffon, 1780. Oiseaux, 7, p. 229). 

d (M. C. Z. 270750) Budongo Forest, U. 29. xi. 38. 

ISPIDINA PICTA NATALENSIS (Smith) 

Alcedo natalensis A. Smith, 1831, S. Afr. Quart. Journ. (1), no. v, p. 14: east of 
Cafferland, i.e. Natal. 

d (M. C. Z. 270751) Magrotto Mtn., T. T. 3. vii. 39. 

Native name. Kumbulu (Kisambara, but not specific). 



PETERS AND LOVERIDGE: AFRICAN BIRDS 237 

Myioceyx lecontei ugandae van Someren 

Myioceyx lecontei ugandae van S6meren, 1921, Bull. Brit. Orn. Club, 41, p. 105: 
Budongo Forest, Uganda. 

cT (M. C. Z. 270752) Budongo Forest, U. 1. xii. 39. 
Breeding. Testes of this topotype very large. 

Halcyon senegalensis cyanoleuca (Vieillot) 
Akedo cyanoleuca Vieillot, 1818, N. Diet. Hist. Nat., 19, p. 401 : Angola. 
d^ 9 9 (M. C. Z. 270753-5) Kitaya, T. T. 27-30. iii. 39. 

Native names. Ngwilile (Kiyao) ; cheleuli (Kimakonde) ; but neither 
specific. 

Halcyon albiventris orientalis Peters 

Halcyon orientalis Peters, 1868, Journ. f. Orn., p. 134: Inhambane, Mozam- 
bique. 

o^ 9 (M. C. Z. 270756-7) Kitaya, T. T. 25 & 30. iii. 39. 
9 (M. C. Z. 270758) Mikindani, T. T. 10. iv. 39. 
d" (M. C. Z. 270759) Lindi, T. T. 2. vi. 39. 
d" (M. C. Z. 270760) Magrotto Mtn., T. T. 8. vii. 39. 

Native names. Libwila (Kimakonde) ; kumhulu (Kisambara, but not 
specific). 

Halcyon chelicuti chelicuti (Stanley) 

Alcedo chelicuti Stanley, 1814, in Salt, Abyssinia, App., p. Ivi: Chelicut, Ethi- 
opia. 

d^ 9 9 (M. C. Z. 270761-63) Ivitaya, T. T. 27-30. iii. 39. 
9 (M. C. Z. 270764) Mbanja, T. T. 27. iv. 39. 
cf (M. C. Z. 270765) Lindi, T. T. 31. v. 39. 

Wings measure: c^d^ 76.7-77.3 mm., 9 9 73.5-78.3 mm. 
Native navies. The same as for H. s. cyanoleucus, in both Kiyao and 
Kimakonde. 

MEROPIDAE 

Melittophagus pusillus meridionalis Sharpe 

Melittophagus meridionalis Sharpe, 1892, Cat. Birds Brit. Mus., 17, p. 45, pi. i, 
fig. 4: Pineto-sv-n, Natal (type in Brit. Mus.). 

o^ 9 9 (M. C. Z. 270766-8) Kitaya, T. T. 25-27. iii. 39. 
d" 9 (M. C. Z. 270769-70) Mikindani, T. T. 19. iv. 39. 
d" 9 9 9 (M. C. Z. 270771-74) Mbanja, T. T. 28. iv. 39. 
9 (M. C. Z. 270775) Lindi, T. T. 31. v. 39. 



238 bulletin: museum of comparative zoology 

Several of these skins are indistinguishable from those of the north- 
eastern form cyanostictus (Cabanis). 

Melittophagus lafresnayii oreobates Sharpe 

Melittophagus oreobates Sharpe, 1892, Ibis, p. 320: Save, Mt. Elgon, i.e. Sabei, 
northern slopes in Uganda {fide Loveridge, 1936). 

9 (M. C. Z. 270776) Mubuku Valley, U. 2. i. 39. 

Native names. Miyoro (Lutoro); musimambugu (Lukonjo). 
Breeding. A large egg forming. 



BUCEROTIDAE 

Bycanistes bucinator bucinator (Temminck) 

Buceros bucinator Temminck, 1824, PL Col. livr. 48, pi. cclxxxiv: Cape of Good 
Hope. 

d' (M. C. Z. 270777) Nchingidi, T. T. 13. v. 39. 

Bycanistes subcylindricus (Sclater) 

Buceros subcylindricus P. L. Sclater, 1870, Proc. Zool. Soc. London, p. 668, 
pi. xxxix: West Africa. 

cf 9 (M. C. Z. 270778-9) Mabira Forest, U. 16. xi. 38. 
Skull of 9 (M. C. Z. ) Budongo Forest, U. 28. xi. 38. 

Bycanistes cristatus brevis Friedmann 

Bycanistes cristatus brevis Friedmann, 1929, Proc. N. Eng. Zool. Club, 11, p. 32: 
Mt. Lutindi, Usambara Mountains, Tanganyika Territory. 

c? (M. C. Z. 270780) Siga Caves, T. T. 14. vi. 39. 

Native name. Hondo (Kisambara). 

LoPHOCEROS melanoleucos suahelicus Neumann 

Lophoceros melanoleucos suahelicus Neumann, 1905, Journ. f. Orn., p. 187: 
Morogoro, Tanganyika Territory. 

cf d^ 9 (M. C. Z. 270781-83) Mikindani, T. T. 12. iv. 39. 
c? cf 9 9 (M. C. Z. 270784-7) Mbanja, T. T. 26-28. iv. 39. 
9 (M. C. Z. 270788) Siga Caves, T. T. 12. vi. 39. 
9 (M. C. Z. 270789) Amboni Estate, T. T. 26. vi. 39. 



PETERS AND LOVERIDGE: AFRICAN BIRDS 239 

Native names. MwiJcwi (Kimakonde, but not specific) ; kwembe 
(Kisambara). 

Breeding. On June 26, ovules small. 

Diet. Stomachs held: (1) Eight green caterpillars; (2) six green 
caterpillars; (3) many grasshoppers; (4) grasshopper legs and green 
stuff; (5) grasshoppers and peanuts; (6) snail and peanuts. 

LoPHOCEROs FASciATUs (Shaw) 
Buceros fasciatus Shaw, 1811, Genl. Zool., 8, pt. i, p. 34: Angola. 
& (M. C. Z. 270790) Budongo Forest, U. 3. xii. 38. 

LoPHOCEROS PALLIDIROSTRIS NEUMANNI ReichcnOW 

Lophoceros nemnanni Peichenow, 1894, Cm. Monatsb., 2, p. 50 (?;om. nud.); 
id. ^'6gel D.-O. Afrikas, p. 128; Mgera, Tanganyika Territory. 

o^ (M. C. Z. 270791) Kitaya, T. T. 25. iii. 39. 

Native names. Likwepe (Kiyao); mwikwi (Kimakonde, but not 
specific). 



CORACIIDAE 

CoRACiAS CAUDATA CAUDATA Linnaeus 
Coracias caudata Linnaeus, 1766, Syst. Nat. ed. 12, 1, p. 160: Angola. 
c? 9 (M. C. Z. 270792-3) Kitaya, T. T. 27. iii. 39. 
Native names. Ngamhwi (Kiyao); gambwila (Kimakonde). 

EuRYSTOMus GLAUCURUS (Miiller) 
Coracias glaucurus P. L. S. Miiller, 1776, Syst. Nat. Suppl., p. 86: Madagascar. 
d' (M. C. Z. 270794) Nchingidi, T. T. 12. v. 39. 

EuRYSTOMUs AFER suAHELicus Neumann 

Eurystomus afer suahelicus Neumann, 1905, Journ. f. Crn., p. 186: "Tschara," 
i.e. Charra, Tana Piver, Kenya Colony. 

c^ (M. C. Z. 270795) Kitaya, T. T. 25. iii. 39. 

Native names. Chole (Kiyao); diole (Kimakonde). 



240 bulletin: museum of comparative zoology 

EuRYSTOMUs gularis neglectus Neumann 

Eurystomus gularis neglectus Neumann, 1908, Orn. Monatsb., 16, p. 28: Can- 
hoca, Angola. 

9 (M. C. Z. 270796) Kibale Forest, U. 10. xii. 38. 

UPUPIDAE 

Upupa africana Bechstein 
Uptipa africana Bechstein, 1811, Kurze Uebers., 4, p. 172: Congo to the Cape 
cf (M. C. Z. 270797) Mbanja, T. T. 28. iv. 39. 

PHOENICULIDAE 

Phoeniculus purpureus marwitzi (Reichenow) 

Irrisor erythrorhynchus marwitzi Reichenow, 1906, Orn. Monatsb., 16, p. 171: 
Mkalama, Kondoa Irangi district, Tanganyika Territory. 

4 c^cf, 1 9 (M. C. Z. 270798-802) Mikindani, T. T. 15. iv. 39. 
o^ (M. C. Z. 270803) Mbanja, T. T. 29. iv. 39. 

Native name. Yelowele (Kimakonde). 

Phoeniculus bollei jacksoni (Sharpe) 

Irrisor jacksoni Sharps, 1890, Ann. Mag. Nat. Hist. (6), 6, p. 503: Kikuyu 
country, Kenya Colony. 

3 c? cf, 2 9 9 (M. C. Z. 270804-8) Idjwi Id., B. C. 27. ii. 39. 

All these birds were shot from the same flock, and near the summit 
of the mountain; their wings measure: 6^6^127-136 mm., 9 9 121- 
131 mm. Thus they compare better with Chapin's (1939, Bull. Am. 
Mus. Nat. Hist., 75, p. 327) measurements of 123-140 mm. for nine- 
teen Congo birds of the montane form, rather than with those of the 
lowland race. 

Rhinopomastus cyanomelas schalowi Neumann 

Rhinopomasttis schalowi Neumann, 1900, Journ. f. Orn., p. 221: Usandawi, 
Tanganyika Territory. 

cf 9 9 (M. C. Z. 270809-11) Kitaya, T. T. 25 & 30. iii. 39. 

d" 9 (M. C. Z. 270812-3) Mikindani, T. T. 17. iv. 39. 

cT (M. C. Z. 270814) Siga Caves, T. T. 8. vi. 39. 



PETERS AND LOVERIDGE: AFRICAN BIRDS 241 

Native navies. Ligolegole (Kiyao); dolowele (Kimakonde at Kitaya); 
nandianhi (Kimakonde at Mikindani). 



CAPITONIDAE 

BuccANODON LEUcoTis KiLiMENSE (Shelley) 

Smilorhis kilimensis Shelley, 1889, Ibis, p. 477: Ivilimanjaro district, Tangan- 
yika Territory. 

cT 9 9 (M. C. Z. 270815-7) Magrotto Mtn., T. T. 30. vi. & 5. vii. 39. 

BuccANODON OLivACEUM wooDWARDi (Shelley) 

Stactolaema woodwardi Shelley, 1895, Bull. Brit. Orn. Club, 5, p. iii: Zululand 
(type in Brit. Mus. from Echowe). 

1 c?' 6 9 9 (M. C. Z. 270818-24) Nchingidi, T. T. 10-12. v. 39. 

These birds have been compared with topotypical woodwardi, 
loaned by the American Museum of Natural History, While they 
exhibit a slight tendency towards oHvaceum, it is not sufficient to 
warrant naming. It is interesting to find woodwardi ranging so far 
north. 

PoGONiuLus LEucoLAiMA MFUMBiRi (Ogilvie-Grant) 

Barbatula mfmnUri Ogilvie-Grant, 1907, Bull. Brit. Orn. Club, 19, p. 107: 
Mfumbiro Volcano, Belgian Ruanda-Urundi. 

& 9 (M. C. Z. 270825-6) Mabira Forest, U. 8. xi. 38. 
9 9 (M. C. Z. 270827-8) Mihunga, U. 11 & 17. i. 39. 

These birds are somewhat intermediate between mfumbiri and 
nyanzae; wings of 9 9 53-54.5 mm. 
Native name. Kaimhakilwa (Lukonjo). 

POGONIULUS BILINEATUS JACKSONI (Sharpc) 

Barbatula jacksoni Sharpe, 1897, Bull. Brit. Orn. Club, 7, p. vii: Alau, Kenya 

Colony. 
Barbatula kandti Reichenow, 1903, Orn. Monatsb., p. 23: Lake Kivu. 

3 cf cf 2 9 9 (M. C. Z. 270829-32) Idjwi Id., B. C. 20-28. ii. 39. 

A special effort was made, at the suggestion of Dr. Chapin, to secure 
this topotypic series of kandti — Kandt lived on Idjwi Id., in Lake 
Khni, — with a view to settling the status of that form. Dr. Chapin 



242 bulletin: museum of comparative zoology 

agrees that he sees no reason for its recognition and has himself (1939, 
Bull. Am. Mus. Nat. Hist., 75, p. 503) followed Grant and Mackworth- 
Praed (1938, Bull. B. O. C, 58, p. 82) in referring it to synonymy. 

Native nmne. Kigofc (Lulega). 

PoGONiULUS BiLiNEATus FiscHERi (Reichenow) 

Barbatula fischeri Eeichenow, 1880, Orn. Centralb., p. 181: No locality (Zanzi- 
bar, vide Journ. f. Crn., 1885, p. 125). 

Pogoniulus bilineatus rovumensis Grote, 1935, Orn. Monatsb., 43, p. 53: Mikin- 
dani, Tanganyika Territory. 

& (M. C. Z. 270833) Kitaya, T. T. 28. iii. 39. 

This bird, with a wing measurement of 51.25 mm., is practically 
topotypic of rovume7isis, Kitaya being on the Rovuma River and only 
twenty miles south of Mikindani. We refer rovumensis to the synony- 
my of fischeri, the wings of our topotypical series of Zanzibar barbets 
being 47.25 to 51 mm. 

INDICATORIDAE 
Indicator indicator (Sparrman) 

Cuculus indicator Sparrman, 1777, Ihil. Trans., 67, p. 43, pi. i: Great Fish 
River, near Somerset East, Cape Frovince. 

9 (M. C. Z. 270834) Mbanja, T. T. 1. v. 39. 

Melignothes exilis exilis Cassin 

Melignothes exilis Cassin, 1856, Froc. Acad. Nat. Sci. Fhiladelphia, p. 157: 
Moonda Eiver, Gaboon. 

cf (M. C. Z. 270835) Idjwi Id., B. C. 20. ii. 39. 

Wing 70 mm. 

Native name. Semanzuki (Lulega). 

PICIDAE 

Campethera nubica scriptoricauda (Reichenow) 

Dendromus scriptoricauda Reichenow, 1896, Crn. Monatsb., 4, p. 131: Bumi, 
Tanganyika Territory. 

9 (M. C. Z. 270836) Kitaya, T. T. 30. iii. 39. 
& (M. C. Z. 270837) Mikindani, T. T. 20. iv. 39. 

Native name. Nyenyena (Kimakonde). 



PETERS AND LOVERIDGE: AFRICAN BIRDS 243 

Campethera cailliautii cailliautii (Malherbe) 

Chrysopicos cailliautii Malherbe, 1849, Rev. Mag. Zool., p. 540: Africa (Mom- 
basa, apud C. Grant. 1915, Ibis, p. 454). 

cT (M. C. Z. 270838) Kitaya, T. T. 30. iii. 39. 
& (M. C. Z. 270839) Nchingidi, T. T. 10. v. 39. 

These have also been compared with Nyasaland fuUeborni, kindly 
lent by Dr. Chapin, but the southern race is a much paler, less greenish, 
form than cailliautii. 

Campethera abingoni suahelica (Reichenow) 

Dendromus chrysurus suahelicus Reichenow, 1902, Vogel Afr., 2, p. 175: Great 
Arusha, Tanganyika Territorj'. 

d" 9 9 (M. C. Z. 270840-42) Kitaya, T. T. 30. iii. 39. 

We tentatively retain this race, despite its recent synonymizing 
with the typical form by Grant and Mackworth-Praed (1939, Bull. 
B.O.C., 60, p. 17) because our long series of skins from Tanganyika 
localities appear to be more definitely yellowish green above than do 
birds from Mt. Selinda, S. Rhodesia. 

Native names. Gongonda (Kiyao); ngongona (Kimakonde). 

Dendropicos fuscescens lepidus (Cabanis & Heine) 

Ipoctonus lepidus Cabanis & Heine, 1863, Mus. Hein., 4, pt. 2, p. 118: Ethiopia. 

9 (M. C. Z. 270843) Idjwi Id., B. C. 2. iii. 39. 

Wing 79 mm. 

Native name. Gongoshera (Lulega). 

Dendropicos fuscescens hartlaubii Malherbe 
Dendropicos hartlaubii Malherbe, 1849, Rev. Mag. Zool., p. 532: Zanzibar. 
3 d'd' (M. C. Z. 270844-6) Kitaya, T. T. 27-30. iii. 39. 

HIRUNDINIDAE 

Hirundo smithii smithii Leach 

Hirundo smithii Leach, App. to Tuckey, 1818, Voy. R. Zaire, p. 407: Chisalla 
Island, lower Congo, Belgian Congo. 

d' 9 (M. C. Z. 270847-8) Kitaya, T. T. 27. iii. 39. 

Native names. Chivalevale '(Kiysio) ; nanyanda (Kimakonde). 



244 bulletin: museum of comparative zoology 

HiRUNDO senegalensis senegalensis Linnaeus 

Hirundo senegalensis Linnaeus, 1766, Syst. Nat. ed. 12, 1, p. 345: Senegal. 

d' 9 (M. C. Z. 270849-50) Mubuku Valley, U. 30. xii. 38. 

Native name. Ndamhera (Lukonjo, but not generic). 
Breeding. These are fledglings. 

PsALiDOPROCNE HOLOMELAENA MASSAicA Neumann 

Psalidoprocne holomelaena massaica Neumann, 1904, Orn. Monatsb., p. 144: 
Ivikuyu, Kenya Colony. 

(f (M. C. Z. 270851) Mubuku Valley, U. 2. i. 39. 
d'd' 9 9 (M. C. Z. 270852-5) Mihunga, U. 11 & 18. i. 39. 

Native names. Ntai (Lutoro); ndambera (Lukonjo, but not generic). 

PSALIDPROCNE PETITI ORIENTALIS ReicheUOW 

Psdlidprocnc orientalis Reichenow, 1889, Journ. f. Orn., p. 277: Lewa, Do- 
doma District, Tanganyika Territory. 

9 (M. C. Z. 270856) Nchingidi, T. T. 11. v. 39. 

Wing 96 mm. ; tail 72 mm. The small size of this bird, as compared 
with cf cf from Kilosa and Tukuyu with wings of 104 and 108 mm., 
may possibly be attributable to immaturity or sex. 

CAMPEPHAGIDAE 

Campephaga flava flava Vieillot 

Canipephaga flava Vieillot, 1817, Nouv. Diet. d'Hist. Nat., 10, p. 49: Southern 
Africa (female). 

d" (M. C. Z. 270857) Idjwi Id., B. C. 27. ii. 39. 
9 (M. C. Z. 270858) Kitaya, T. T. 5. iv. 39. 

Native names. Kikliku (Lulega); nankahichi (Kimakonde). 

CoRACiNA CAESIA PURA (Sharpc) 

Graucalus purus Sharpe, 1891, Ibis, p. 121: Mt. Elgon. 

9 (M. C. Z. 270859) Mubuku Valley, U. 4. i. 39. 

9 (M. C. Z. 270860) Mihunga Ridge, U. 16. i. 39. 

& (M. C. Z. 270861) Magrotto Mtn., T. T. 10. vii. 39. 

The male shows a tendency towards typical caesia of South Africa. 



PETERS AND LOVERIDGE : AFRICAN BIRDS 245 

Native names. Kaneju (Lutoro); Idnihioe (Lukonjo). 
Breeding. In both females from the Ruwenzori Mountains the ova 
are greatly enlarged and an egg formed. 

DICRURIDAE 

DiCRURUS ADSiMiLis DiVARiCATUs (Lichtenstein) 

Musicapa divaricata Lichtenstein, 1823, Verz. Doubl. Zool. Mus. Berlin, p. 52: 
Senegal. 

& cf (M. C. Z. 270862-3) Kitaya, T. T. 25 & 27. ui. 39. 
d^ d^ 9 (M. C. Z. 270864-6) Mikindani, T. T. 17. iv. 39. 
d^ d^ 9 9 (M. C. Z. 270867-70) Mbanja, T. T. 29. iv. 39. 

Xative navies. Lihwilili (Kiyao) ; hiiduicilu (Kimakonde at Kitaya) ; 
namha (Kimakonde at Mikindani). 

DiCRURUS LUDWiGii LUDWiGii (Smith) 

Edolius ludicigii A. Smith, 1834, S. Afr. Quart. Journ. (2), p. 144: Port Natal 
i.e. Durban, Natal. 

d" 9 (M. C. Z. 270871-2) Nchingidi, T. T. 11-12. v. 39. 
d' (M. C. Z. 270873) Magrotto Mtn., T. T. 6. vii. 39. 

X afire name. Mlamha (Kisambara). 

ORIOLIDAE 

Oriolus monacha rolleti Salvadori 

Oriolus rolleti Salvadori, 1864, Atti Soc. Italiana Sci: Nat. Milano, 7, Riunione 
a Biella, p. 161 : WTiite Nile between Lat 4° and 5° N. 

cf (M. C. Z. 270874) Mabira Forest, U. 11. xi. 38. 
4 d' d' (M. C. Z. 270875-6) Idjwi Id., B. C. 23-28. ii. 39. 

Wings 124, 124.5, 130, 131, and 131 mm. respectively. 
Native name. Lushauiinga (Lulega). 

Breeding. On November 11, at IVIubango, Mabira Forest, testes 
large. 

Oriolus monacha kikuyuensis van Someren 

Oriolus larvatus kikuyuensis van Someren, 1922, Nov. Zool., 29, p. 127: Nairobi, 
Kenya Colony. 

d^ (M. C. Z. 270877) Kitaya, T. T. 28. iii. 39. 
d' (M. C. Z. 270878) Mikindani, T. T. 14. iv. 39. 



246 bulletin: museum of comparative zoology 

The Mikindani bird is moulting, but the wing of the Kitaya oriole 
is 133 mm., so they are referred tentatively to kikuyuensis rather than 
to reichenowi, which has been reported on the coast to the north. 

Native names. Likudo (Kiyao); likuhwamhu (Kimakonde at Kit- 
aya); lichendi (Kimakonde at Mikindani). 

Oriolus chlorocephalus Shelley 

Oriolus chlorocephalus Shelley, 1896, Ibis, p. 183, pi. iv: Mt. Chiradzulu, Nyasa- 
land. 

c^cj^ (M. C. Z. 270879-80) Nchingidi, T. T. 11-12. v. 39. 

The finding of this beautiful oriole on the Rondo Plateau helps to 
bridge a gap in its spotty distribution. 



PARIDAE 

Parus funereus (Verreaux and Verreaux) 
Melanoparus funereus J. & E. Verreaux, 1855, Journ. f. Orn., p. 104: Gaboon. 
cf 9 (M. C. Z. 270881-2) Kibale Forest, U. 12. xii. 38. 

Parus fasciiventer Reichenow 

Parus fasciiventer Reichenow, 1893, Orn. Monatsb., 1, p. 31: Ruwenzori Mtns. 

cf (M. C. Z. 270883) Mubuku Valley, U. 30. xii. 38. 
d^ (M. C. Z. 270884) Mihunga Ridge, U. 10. i. 39. 

Native name. Kakomogoli (Lukonjo). 
Breeding. Of these topotypes, the December bird is a nestling; 
the January bird was bald when shot. 



TIMALIIDAE 

TURDOIDES PLEBEJA KIRKI (Sharpc) 

Crateropus Mrki Sharpe, 1876, in Layard, Birds S. Africa, ed. 2, p. 213: Zam- 
bezi country (type in Brit. Mus. from Mazzaro). 

cf 9 (M. C. Z. 270885-6) Kitaya, T. T. 30. iii. 39. 

Native names. Likusesi (Kiyao); y ay ay a (Kimakonde). 



PETERS AND LOVERIDGE: AFRICAN BIRDS 247 

Illadopsis pyrrhopterus (Reichenow & Neumann) 

Callene pyrrhoptcra Eeichenow & Neumann, 1895, Orn. Monatsb., 3, p. 75: 
Mau, Kenya Colony. 

d" (M. C. Z. 270887J Mihunga Ridge, U. 13. i. 39. 

This species does not seem to have found its final systematic resting 
place. It was originally described in 1895 as a Callene and was still 
carried in this genus of Turdidae in Sharpe's Handlist. When Banner- 
man named the TimaHne genus Pseudoalcippe in 1923, this was one 
of the species transferred to it. Sclater, in 1930, followed Bannerman 
in his treatment of purrhopterus, but van Someren, in 1932, disagreed, 
and insisted that the bird was an Illadopsis; Friedmann & Loveridge, 
1937, still included it in Pseudoalcippe. 

The chief characters separating Pseudoalcippe from Illadopsis are 
the more sharply defined color pattern, weaker feet and bill, and 
shorter rictal bristles of the former. Pyrrhopterus does not fit exactly 
into either, occupying a more or less intermediate position, but perhaps 
nearest to Illadopsis in respect to bill and rictal bristles; further study 
may result in the union of the two genera. 

Native name. Kiniva (Lukonjo). 

Macrosphenus flavicans hypochondriacus (Reichenow) 

Rectirostrum hypochondriacum Reichenow, 1893-, Orn. Monatsb., 1, p. 32: Kin- 
jawanga, near Beni, Belgian Congo. 

9 (M. C. Z. 270888) Budongo Forest, U. 2. xii. 38. 



PYCNONOTIDAE 

Pycnonotus tricolor minor Heuglin 

Pycnonotus nigricans var. minor Heuglin, 1869, Orn. Nordost. Afr., 1, p. 398: 
Bahr el Abiad, i.e. Upper White Nile. Anglo-Egyptian Sudan. 

c? (M. C. Z. 270889) Mabira Forest, U. 14. xi. 38. 
2 c^cT (M. C. Z. 270890-1) Mihunga Ridge, U. 11-28. i. 39. 

(^ (M. C. Z. 270892) Nyakabande, U. 28. i. 39. 
cf cf 9 (M. C. Z. 270893-4) Idjwi Id., B. C. 17-20. ii. 39. 

Wings of d'c^, 89.5, 92, 93.5 and 93.5 mm., 9 9 , 93.5, 94 and 99 
mm. It should be borne in mind that the sexing of these birds was 
done by a native, for, while apparently minor, two of the males (from 



248 bulletin: museum of comparative zoology 

Mihunga and Idjwi) are slightly below the figures furnished by van 
Someren, viz. cfd", 93-99 mm., 9 9 , 88-94 mm. 

Native names. Sosolye (Luganda) ; isole (Lutoro and Lukonjo) ; 
lusholea (Lulega). 

Breeding. On November 14, testes large. 

Pycnonotus tricolor micrus Oberholser 

Pycnonotus layardi micrus Oberholser, 1905, Proc. U. S. Nat. Mus.. 28, p. 891: 
Taveta, Kenya. Colony. 

cf (M. C. Z. 270895) Kitaya, T. T. 25. iii. 39. 

d" (M. C. Z. 270896) Mbanja, T. T. 1. v. 39. 

9 (M. C. Z. 270897) Nchingidi, T. T. 10. v. 39. 

9 (M. C. Z. 270898) Magrotto Mtn., T. T. 4. vii. 39. 

Wings of Kitaya d^, 90 mm., but worn; Mbanja cf molting; Nchin- 
gidi 9 . 86 mm. ; Magrotto 9 , 85 mm. 

Native names. Lisolokoto (Kiyao); namtindi (Kimakonde); choe 
(Kisambara). 

Atimastillus flavicollis pallidigula (Sharpe) 

Xenocichla pallidigula Sharpe, 1897, Bull. Brit. Orn. Club. 7, p. 7: Entebbe, 
Uganda. 

d" (M. C. Z. 270899) Idjwi Id., B. C. 3. iii. 39. 

Native name. Kichwagashwaga (Lulega). 

Phyllastrephus debilis debilis (Sclater) 

Xenocichla debilis W. Sclater, 1899, Ibis, p. 284: Near Inhambane, Mozambique 
cT (M. C. Z. 270900) Nchingidi, T. T. 11. v. 39. 

Phyllastrephus flavostriatus tenuirostris 

(Fischer & Reichenow) 

Xenocichla tenuirostris Fischer & Reichenow, 1884, Journ. f. Orn., p. 262: Lindi, 
Tanganyika Territory. 

d^ (? (M. C. Z. 270901-2) Nchingidi, T. T. 11. v. 39. 
? 9 (M. C. Z. 270903) Magrotto Mtn., T. T. 12. vii. 39. 



PETERS AND LOVERIDGE: AFRICAN BIRDS 249 

The Nchingidi males are almost topotypic, and Chapin, who has 
kindly examined them, thinks that the unsexed Magrotto bird — 
despite its smaller size, wing 82 mm. — should also be referred to 
tenuirosfris. This race differs from the typical form in being somewhat 
lighter and more plainly marked with yellow below. The recently 
described race vinccnti C. Grant (1940, Bull. Brit. Orn. Club, 60, p. 
62) from Namuli Mtns., and southern Nyasaland, is but slightly 
different. 



Arizelocichla tephrolaema kikutuensis (Sharpe) 

Xenocichla kikuyuensis Sharpe, 1891, Ibis, p. 118: Ivikuyu, Kenya Colony. 

d^ ? 9 (M. C. Z. 270904-6) Mubuku ^'alley, U. 30. xii. 38-5. i. 39, 

Native name. Kiwota (Lutoro and Lukonjo). 
Breeding. On January 2, 1 large egg forming. 



Arizelocichla milanjensis striifacies (Reichenow & Neumann) 

Xenocichla striifacies Eeichenow & Neumann, 1895, Orn. Monatsb., 3, p. 74, 
Marangu, Mt. Kilimanjaro, Tanganyika Territory. 

c^ 9 (M. C. Z. 270907-8) Magrotto Mtn., T. T. 29. vi. 39. 

Native name. Kojojo (Kisambara). 
Breeding. On June 29, a large ovule. 

Arizelocichla masukuensis roehli (Reichenow) 

Andropadus roehli Reichenow, 1905, Orn. Monatsb., 13, p. 181: Mlalo, Usam- 
bara Mountains, Tanganyika Territory. 

9 (M. C. Z. 270909) Magrotto Mtn., T. T. 7. vii. 39. 

Charitillas gracilis ugandae (van Someren) 

Andropadus ugandae van Someren, 1915, Bull. Brit. Orn. Club, 35, p. 127: 
Mabira Forest, Uganda. 

9 9 (M. C. Z. 270910-11) Budongo Forest, U. 1. xii. 38. 

In the absence of adequate comparative material, we employ van 
Someren's name as being undoubtedly applicable, but without vouch- 
ing for its distincteness from typical gracilis of Angola. 



250 bulletin: museum of comparative zoology 

Stelgidocichla latirostris EUGENIA (Rcichenow) 

Andropadus eugenius Reichenow, 1892, Journ. f. Orn., p. 53: Bukoba, Tangan- 
jaka Territory. 

3 d'd' 1 9 (M. C. Z. 270912-4) Idjwi Id., B. C. 20-21. ii. 39. 
Native name. Lugwate (Lulega). 

Eurillas virens holochlorus van Someren 

Eurillas virens holochlorus van Someren, 1922, Nov. ZooL, 29, p. 189: Sezibwa, 
Uganda. 

cJ' (M. C. Z. 270915) Budongo Forest, U. 1. xii. 38. 
Breeding. Grass in bill when shot. 

Eurillas virens marwitzi (Reichenow) 

Andropadus marwitzi Reichenow, 1895, Orn. Monatsb., 3, p. 188: Marangu, 
Mt. Kilimanjaro, Tanganyika Territory. 

9 (M. C. Z. 270916) Nchingidi, T. T. 11. v. 39. 

9 (M. C. Z. 270917) Magrotto Mtn., T. T. 3. vii. 39. 

Native name. Kojojo (Kisambara, but not generic). 



TURDIDAE 

TURDUS LIBONYANUS COSTAE Rcnsch 

Turdus libonyanus costae Rensch, 1923, Journ. f. Orn., p. 99: Magogoni, lower 
Ruvu River, Tanganyika Territory. 

d" (M. C. Z. 270918) Lindi, T. T. 31. v. 39. 

Turdus olivaceus baraka (Sharpe) 

Merula baraka Sharpe, 1903, Bull. Brit. Orn. Club, 14, p. 19: Ruwenzori Mtns., 
Uganda. 

(? 9 (M. C. Z. 270919-20) Mihunga Ridge, U. 10 & 12. i. 39. 

Native names. Kinyabutoke (Lutoro); mhuhe (Lukonjo). 
Breeding. On January 12, the female was feeding a fledgling. 



PETERS AND LOVERIDGE : AFRICAN BIRDS 251 

Neocossyphus rufus arrhenii Lonnberg 

Neocossyphus rufus arrhenii Lonnberg, 1917, Arkiv Zool., 10, No. 24, p. 30: 
Beni, Semliki Valley, Belgian Congo. 

? (M. C. Z. 270921) Budongo Forest, U. 24. xi. 38. 

Saxicola torquata axillaris (Shelley) 

Pratincola axillaris Shelley, 1884, Proc. Zool. Soc. London, p. 556: Mt. Kili- 
manjaro, 7000 feet, Tanganyika Territory. 

juv. cf (M. C. Z. 270922) Mushongero, U. 3. ii. 39. 
? c? (M. C. Z. 270923) Idjwi Id., B. C. 21. ii. 39. 

Native names. Fundi (Lugezi); kashaveye (Lulega). 

Erythropygia hartlaubi Reiehenow 

Erythropygia hartlaubi Reiehenow, 1891, Journ. f. Orn.. p. 63: Mutjora, Sem- 
liki Valley (?Mtsora of Stanley). 

cf (M. C. Z. 270924) Mihunga Ridge, U. 11. i. 39. 
N^ alive name. Enumbi (Lukonjo). 

POGONOCICHLA STELLATA INTENSA Sharpe 

Pogonocichla intensa Sharpe, 1901, Bull. Brit. Orn. Club, 11, p. 67: Ntebi, i.e. 

Entebbe, L^ganda. 
Tarsiger ruwenzorii Cgilvie-Grant, 1906, Bull. Brit. Orn. Club, 19, p. 33: 

Mubuku \'alley, 7000 feet. Fuwenzori Mtns., Uganda. 

cf cT 9 ? (M. C. Z. 270925-8) Mubuku Valley, U. 31. xii. 38 & 3. i. 39 

These specimens are topotypical of ruwenzorii, but this name is 
currently synonymized with intensa; lacking topotypical material of 
the latter we are not able to express an opinion as to the correctness of 
this course. We can detect no essential diflferences between this series 
and two males from the forest along the Lulenga-Kivu Road, eastern 
Congo, on the one hand, and a male from Ruanda on the other. 

Native name. Nombi (Lukonjo). 

SYLVIIDAE 

Sylvia borin (Boddaert) 

Motacilla borin Boddaert, 1783, Table PL Enl., p. 35: France {ex Daubenton, 
PI. Enl., p. 579, fig. 2). 

9 (M. C. Z. 270929) Mabira Forest, U. 12. xi. 38. 

On migration. 



252 bulletin: museum of comparative zoology 

Seicercus laetus (Sharpe) 

Cryptolopha laeta Sharpe, 1902, Bull. Brit. Orn. Club, 13, p. 9: Ruwenzori 
Mountains, Uganda. 

d^ c? 9 ? (M. C. Z. 270930-33) Mubuku Valley. U. 31. xii. 38-7. i. 39. 

Native name. Kaboliamaisu (Lukonjo). 

Apalis flavida neglecta (Alexander) 

Chlorodyta neglecta Alexander, 1900, Bull. Brit. Orn. Club, 10, p. 17: S. E. 
Africa (Type in Brit. Mus. from Zambezi River). 

d' 9 (M. C. Z. 270934-5) Kitaya, T. T. 25 & 27. iii. 39. 
cf 9 (M. C. Z. 270936-7) Nchingidi, T. T. 10. v. 39. 

Native name. Chitakaka (Kimakonde). 

Apalis binotata personata Sharpe 

Apalis personata Sharpe, 1902, Bull. Brit. Orn. Club, 13, p. 9: Ruwenzori 
Mountains, Uganda. 

9 (M. C. Z. 270938) Mihunga Ridge, U. 12. i. 39. 

Native name. Kasandangaleka (Lukonjo). 

Apalis eidos sp. nov. 

4 cf d^ 4 9 9 (M. C. Z. 270939-44) Idjwi Id., B. C. 17. ii-1. iii. 39. 

Type. Museum of Comparative Zoology, No. 270942, an adult male 
from Upper Mulinga River, Idjwi Island, Lake Kivu, Belgian Congo, 
collected by Arthur Loveridge, February 28, 1939. 

Description. Allied to A. argcnta Moreau (1941, Bull. Brit. Orn. 
Club, 61, No. 437, p. 47) from Kungwe Mountain, western Tanganyika 
Territory, but, according to Dr. Chapin, who has compared our birds 
with the type of argenta, specifically distinct. 

It differs from argenta in lacking the narrow silver edges on the inner 
primaries; in having the upper parts, except the crown, with a distinct 
olive green wash; in having the underparts white, shading to pale 
gray on sides of breast and to greenish olive on the flanks; and the 
underwing-coverts pale yellowish. 

Measurements. Wings of cfcf , 46 (Type), 47.5, 48 and 48 mm.; of 
9 9 , 44.5, 44.5, 46, and 46.5 mm. Tails of d' d', 47 (Type), 47, 48 mm. 
and one damaged; of 9 9 , 43, 45, 45.5 mm. and one damaged. 



PETERS AND LOVERIDGE: AFRICAN BIRDS 253 

Remarks. A pair of these birds have been donated to the Congo 
Museum, Tervueren. For description of type locality see final report 
of this series. 

Native name. Kifusambiro (Lulega). 

Apalis ruwenzorii ruwenzorii Jackson 

Apalis ruwenzorii Jackson, 1904, Bull. Brit. Orn. Club, 15, p. 11: Ruwenzori 
Mountains, Uganda. 

c^ (M. C. Z. 270945) Mubuku Valley, U. 31. xii. 38. 
Native name. Kaswebe (Lukonjo). 

Sylvietta whytii whytii (Shelley) 

Sylviella [sic] whytii Shelley, 1894, Ibis, p. 13: Zomba, Nyasaland. 
Sylviella whytei [sic] pallidior Grote, 1911. Orn. Monatsb., 19, p. 163: Mikin- 
dani, Tanganyika Territory. 

o^ cf 9 9 (M. C. Z. 270946-9) Kitaya, T. T. 28. iii. & 3. iv. 39. 

Kitaya is scarcely twenty miles south of the type locality of palli- 
dior; unfortunately we have no topotypical whytii for comparison. 
Three of these birds were shot from one flock, yet two (cf 9 ) are 
darker and two (cf 9 ) are paler, the latter comparing well with birds 
from Dar es Salaam, while the darkest of the series are almost as dark 
as the race jacksom from Kamasia, Kenya Colony. 

Eremomela scotops occipitalis (Fischer & Reichenow) 

Tricholais occipitalis Fischer & Reichenow, 1884, Journ. f. Orn., p. 181: Pan- 
gani River, Tanganyika Territory. 

d^ c^ 9 (M. C. Z. 270951-53) Kitaya, T. T. 30. iii. 39. 

Native name. Jijiliwa (Kimakonde). 

Camaroptera brachyura littoralis Grote 

Camaroptera pileata littoralis Grote. 1911, Orn. Monatsb., 19, p. 163: Mikin- 
dani, Tanganyika Territory. 

c? (M. C. Z. 270950) Kitaya. T. T. 3. iv. 39. 

In referring this almost topotypic bird to littoralis, we follow Fried- 
mann (1937, Bull. Mus. Comp. Zool., 81, p. 269) in using the name, 
for we lack good comparative material. 

Native names. Katinyc (Kiyao); mdinye (Kimakonde). 



254 bulletin: museum of comparative zoology 

Camaroptera brevicaudata griseigula Sharpe 

Camaroptera griseigula Sharpe, 1892, Ibis, p. 158: Voi River, Teita, Kenya 
Colony. 

juv. 9 (M. C. Z. 270954) Mabira Forest, U. 18. xi. 38. 
c^ cf (M. C. Z. 270955-6) Idjvvi Id., B. C. 17 & 24. ii. 39. 

Native name. Ogosashalcwe (Lulega). 



Camaroptera superciliaris ugandae Clarke 

Camaroptera superciliaris ugandae S. Clarke, 1914, Bull. Brit. Orn. Club, 33, 
p. 136: Uganda. 

9 (M. C. Z. 270957) Mabira Forest, U. 15. xi. 38. 
9 (M. C. Z. 270958) Budongo Forest, II. 29. xi. 38. 

Breeding. On November 15, in breeding condition. 



Cisticola chiniana heterophrys Oberholser 

Cisticola heterophrys Cberholser, 1906, Ann. Carnegie Mus., 3, p. 496: Mom- 
basa, Kenya Colony. 

d" 9 (M. C. Z. 270959-60) Magrotto Mtn., T. T. 3 & 5. vii. 39. 

Recently recorded by Sclater & Moreau (1933, Ibis, p. 28) from the 
nearby Usambara Mountains. 

Native name. Mkucha (Kisambara, but not specific). 



Cisticola chubbi Sharpe 

Cisticola chubbi Sharpe, 1892, Ibis, p. 157: Kimangitchi, Mt. Elgon. 

cf c^ 9 (M. C. Z. 270961-3) Mihunga Ridge, U. 12 & 18. i. 39. 

Cisticola erythrops sylvia Reichenow 

Cisticola sylvia Feichenow, 1904, Orn. Monatsb., p. 28: Ulegga, i.e. Warega, 
near Lake Albert. 

9 (M. C. Z. 270964) Idjwi Id., B. C. 3. iii. 39. 

Native name. Kifusambilo (Lulega). 



PETERS AND LOVERIDGE: AFRICAN BIRDS 255 

CiSTICOLA ERYTHROPS NYASA LynCS 

Cisticola erythrops nyasa Lynes, 1931, Ibis, p. 374: Chiromo, Ruo district, 
Nyasaland. 

o^ (M. C. Z. 270965) Kitaya, T. T. 3. iv. 39. 
Native names. Lichonja (Kiyao); ndumhwi (Kimakonde). 



Prinia mistacea immutabilis van Someren 

Prinia mistacea immutabilis van Someren, 1920, Bull. Brit. Orn. Club, 40, p. 
93: Lake Nakuru, Kenya Colony. 

cf" (M. C. Z. 270966) Mabira Forest, U. 8. xi. 38. 
d^ 9 (M. C. Z. 270967-8) Budongo Forest, U. 2-3. xii. 38. 
d^ (M. C. Z. 270969) Bundibugyo, U. 20. xii. 38. 

Native names. Ndiahununu (Lutoro); sesi (Luamba). 



Prinia mistacea tenella (Cabanis) 

Drymoica tenella Cabanis, 1869, in von der Decken, Reise, 3, p. 23: Mombasa, 
Kenya ColonJ^ 

cf c^ 9 9 (M. C. Z. 270970-73) Kitaya, T. T. 25. iii.-3. iv. 39. 

(f (M. C. Z. 270974) Mikindani, T. T. 19. iv. 39. 

d' 9 (M. C. Z. 270975-6) Mbanja, T. T. 26. iv. 39. 

cf (M. C. Z. 270977) Nchingidi, T. T. 10. v. 39. 

9 9 (M. C. Z. 270978-9) Lindi, T. T. 31. v. 39. 

d" (M. C. Z. 270980) Siga Caves, T. T. 15. vi. 39. 

d (M. C. Z. 270981) Magrotto Mtn., T. T. 13. vii. 39. 

The Lindi bird's tail was in process of moulting when shot. 
Native names. Kandidi (Kiyao) ; mchapi (Kimakonde) ; mkucha 
(Kisambara, but not generic). 



Prinia leucopogon reichenowi (Hartlaub) 

Burnesia reichenowi Hartlaub, 1890, Journ. f. Orn., p. 151: Njangalo, north- 
eastern Belgian Congo. 

d (M. C. Z. 270982) Mihunga Ridge, U. 12. i. 39. 



256 bulletin: musei^m of comparative zoology 

MUSCICAPIDAE 
Alseonax minimus pumilus Reichenow 

Alseonax pumila Reichenow, 1892, Journ. f. Orn., pp. 32, 218: Bukoba, Tan- 
ganyika Territory. 

cf (M. C. Z. 270983) Mabira Forest, U. 15. xi. 38. 
.9 (M. C. Z. 270984) Mubuku Valley, U. 3. i. 39. 

Native name. Ndonairi (Lukonjo). 

Alseonax minimus subtilis Grote 

Alseonax murinus subtilus Grote, 1920, Orn. Monatsb., 28, p. 114: Beni, Bel- 
gian Congo. 

2 c? cf , 3 9 9 (M. C. Z. 270985-8) Idjwi Id., B. C. 1. iii. 39. 

Native name. Torotonzi (Lulega). 

Alseonax aquaticus ruandae Gyldenstolpe 

Alseonax infulatus ruandae Gyldenstolpe, 1922, Bull. Brit. Orn. Club, 43. p. 36: 
Bufundi, Kigezi district, Uganda. 

c? 9 (M. C. Z. 270989-90) Mushongero, U. 1. ii. 39. 

Native name. Kasindikera (Lugezi). 

Breeding. Another of these birds was observed building. 

Alseonax flavipes Bates 

Alseonax flavipes Bates, 1911, Ibis, p. 522: Camma River, Gaboon. 
Pedilorhynchus cpulatus sethsmithi van Someren, 1922, Nov. Zool., 29, p. 96: 
Budongo Forest, Uganda. 

9 (M. C. Z. 270991) Budongo Forest, U. 25. xi. 38. 
A topotype of sethsmithi. 

Bradornis pallidus murinus Hartlaub & Finsch 

Bradyornis [sic] murinus Hartlaub & Finsch, in Finsch & Hartlaub, 1870, 
Vogel Ost-Afr., Nachtr. p. 866: Caconda, Angola. 

9 9 (M. C. Z. 270992-93) Mbanja, T. T. 26. iv. & 3. v. 39. 



PETERS AND LOVERIDGE: AFRICAN BIRDS 257 

DiOPTRORNis TOROENSis (Hartert) 

Muscicapa taroensis Hartert, 1900, Novit. Zool., 7, p. 37: Fort Gerry, i.e. Fort 
Portal, Uganda. 

& (M.C.Z. 270994) IMihunga Ridge, U. 12. i. 39. 
9 (M.C.Z. 270995) Idjwi Id., B. C. 18. ii. 39. 

Native names. Kimva (Lukonjo); shanshaluke (Lulega), 



Melaenornis pammelaina tropicalis (Cabanis) 

Melanopepla tropicalis Cabanis, 1884, Journ. f. Orn., p. 241: Ikanga, Ukafnba, 
Kenya Colony. 

9 9 (M. C. Z. 270996-7) Kitaya, T. T. 25. iii. 39. 
9 9 (M. C. Z. 270998-9) Mikindani, T. T. 15. iv. 39. 
9 (M. C. Z. 270000) Lindi, T. T. 31. v. 39. 

The wings of these five females range from 96.5 to 99 mm., thus 
agreeing with Friedmann's (1937, U. S. Nat. Mus., Bull. 153, p. 233) 
findings that the northern race extends southwards to central Moz- 
ambique. 

Native names. Namba (Kiyao and Kimakonde at Kitaya); liulu- 
wilu (Kimakonde at Mikindani). 



Megabyas flammulatus aequatorialis Jackson 

Megahias [sic\ aequatorialis Jackson, 1904, Bull. Brit. Orn. Club, 15, p. 11: 
Entebbe, Uganda. 

d" (M. C. Z. 271001) Kibale Forest, U. 13. xii. 38. 



Bias musicus changamwensis van Someren 

Bias musictis changamwensis van Someren, 1919, Bull. Brit. Orn. Club, 40, 
p. 24: Changamwe, near Mombasa, Kenya Colony. 

9 (M. C. Z. 271002) Kitaya, T. T. 3. iv. 39. 

cf (M. C. Z. 271003) Mikindani, T. T. 17. iv. 39. 

This appears to be a somewhat doubtful form, our single female is 
perhaps, slightly paler than a topotypical feminina. 

Native name. Nandiendiende (Kimakonde at Mikindani). 



258 bulletin: museum of comparative zoology 

Batis reichenowi Grote 

Batis reichenowi Grote, 1911, Orn. Monatsb., 19, p. 162: Mikindani, Tangan- 
zika Territory. 

9 (M. C. Z. 271004) Nchingidi, T. T. 11. v. 39. 

Rondo Plateau, where this bird was shot, is almost due west of the 
type locality. Dr. Chapin concurs in the identification, pointing out 
that it diflFers only from the colored plate (1912, Journ. f. Orn., pi. viii) 
of reichenowi in the figured type having a little rufous mixed with the 
gray chest band as well as on the flanks. 

Batis mixta (Shelley) 

Pachyprora mixta Shelley, 1889, Proc. Zool. Soc. London, p. 359, pi. xl: Mt. 
Ivilimanjaro, 6000-7000 feet, Tanganyika Territory. 

c? 9 (M. C. Z. 271005-6) Magrotto Mtn., T. T. 8. vii. 39. 

Native name, Kikoda (Kisambara). 

Batis molitor soror Reichenow 

Batis puella soror Reichenow, 1903, Vogel Afr., 2, p. 485: Quilimane, Mozam- 
bique. 

9 (M. C. Z. 271007) Kitaya, T. T. 30. iii. 39. 
cTcf 9 9 (M. C. Z. 271008-11) Mbanja, T. T. 27. iv. 39. 
9 (M. C. Z. 271012) Nchingidi, T. T. 10. v. 39. 

Native names. Galokotoio (Kiyao); chikakodongo (Kimakonde). 

Batis minor nyansae Neumann 

Batis minor nyansae Neumann, 1907, Journ. f. Orn., p. 354: Kiva Mtesa, 
Uganda. 

c^ (M. C. Z. 271013) Mihunga Eidge, U. 12. i. 39. 

Native names. Narukaka (Lutoro); kasaramaganda (Lukonjo). 

Platysteira peltata peltata Sundevall 

Platysteira peltata Sundevall, 1850, Oefv. Vet.-Ak. Forh., 7, p. 105: "CaflFraria 
Inferiore," i.e. Natal. 

? 9 (M. C. Z. 271014) Kitaya, T. T. 29. iii. 39. 
Native names. Namoto (Kiyao); mkwidimbuladingope (Kimakonde). 



PETERS AND LOVERIDGE: AFRICAN BIRDS 259 

Platysteira cyanea nyansae Neumann 

Platysteira cyanea mjansae Neumann, 1905, Journ. f. Orn., p. 210: Bukoba, 
Tanganyika Territory. 

9 (M. C. Z. 271015) Idjwi Id., B. C. 2. iii. 39. 
Native name. Yongoyongo (Lulega). 

Erythrocercus livingstonei thomsoni Shelley 

Erythrocercus tJxomsoni Shelley, 1882, Proc. Zool. Soc. London, p. 303, pi. xvi» 
fig. 2: Rovuma River. 

cf (M. C. Z. 271016) Nchingidi, T. T. 10. v. 39. 

We are indebted to Dr. Chapin for the identification of this bird. 
He points out that Vincent found intermediates between livingstonei 
Gray of the Zambezi, and thomsoni Shelley of the Rovuma, in northern 
Mozambique, viz. vionapo Vincent (1933, Bull. Brit. Orn. Club, 53, 
p. 137), later (1934, Ibis, pp. 390-3, pi. xiv) discussing and figuring 
it. He considers the genus Chloropetella inseparable from Erythro- 
cercus. 

Erranornis longicauda teresita (Antinori) 

Elminia teresita Antinori, 1864, Cat. Uccelli, p. 50: Djur, Bahr el Ghazali 
Anglo-Egj'ptian Sudan. 

? (M. C. Z. 271017) Mihunga Ridge, T. T. 14. i. 39. 
Native name. Kasanzamatekere (Lukonjo). 

Trochocercus albonotatus albonotatus Sharpe 
Trochocercus albonotatus Sharpe, 1891, Ibis, p. 121: Mt. Elgon. 

d' 9 9 (M. C. Z. 271018-20) Mubuku Valley, U. 30-31. xii. 38. 
Native name. Kasihiraho (Lukonjo). 



MOTACILLIDAE 

MOTACILLA AGUIMP VIDUA Suudcvall 

Motacilla vidim Sundevall, 1850. Oefv. Vet.-Ak. Forh., 7, p. 128: (Type in 
Stockholm Museum from Syene, i.e. Assouan, Upper Egypt). 

cf cf 9 (M. C. Z. 271021-22) Mushongero, U. 1. ii. 39. 



260 bulletin: museum of comparative zoology 

One male is in juvenile plumage, the three birds were a family party. 
Native name. Nyamanza (Lugezi, but not specific). 

Motacilla CLARA Sharpc 

Motacilla clara Sharpe, 1908, Ibis, p. 341 : nom. nov. pro. M. longicauda Riippell 
(not Gmelin), 1840, Neue Wirbelth., p. 84, pi. xxix, fig. 2: Simen, Ethiopia. 

9 (M. C. Z. 271023) Magrotto Mtn., T. T. 5. vii. 39. 

Native name. Kiwiwimazi (Kisambara). 

Motacilla capensis wellsi Ogilvie-Grant 

Motacilla ivellsi Ogilvie-Grant, 1911, Bull. Brit. Orn. Club, 29, p. 30: Kigezi, 
S. W. Uganda. 

d" (M. C. Z. 271024) Mushongero, U. 1. ii. 39. 

This bird is topotypic, Mushongero being on Lake Mutanda. 
Kigezi. Its Lugezi name is the same as that for vidua, given above. 

Budytes flavus thunbergi (Billberg) 

Motacilla thxmhergi Billberg, 1828, Syn. Faun. Scand., 1, pt. 2, Aves, p. 50: 
Lapland. 

d" 9 (M. C. Z. 271025-6) Budongo Forest, U. 1. xii. 38. 
(d (M. C. Z. 271027) Mushongero, U. 3. ii. 39. 

Both Budongo migrants were sexed as female by the native skinner, 
one, however, appears to be a male. 

Native name. Nyamonyo (Lugezi). 

Anthus sordidus nyassae Neumann 

Anthus nicholsoni nyassae Neumann, 1906, Journ. f. Orn., p. 233: between 
Sangesi and Songea, Tanganyika Territory. 

3 d^ c^ 2 9 9 (M. C. Z. 271028-32) Mbanja, T. T. 26-27. iv. 39. 

Macronyx sharpei Jackson 

Macronyx sharpei Jackson, 1904, Bull. Brit. Orn. Club, 14, p. 74: Mau Plateau, 
Kenya Colony. 

9 (M. C. Z. 271033) S. Kinangop, K. C. 28. x. 38. 

This, the only Kenya bird in the collection, was non-breeding. We 
are indebted to Sir Charles Belcher for it. 



PETERS AND LOVERIDGE: AFRICAN BIRDS 261 

LANIIDAE 

Lanius mackinnoni Sharpe 

Lanius mackinnoni Sharpe, 1891, Ibis, pp. 444, 596, pi. xiii: Kikuyu, Kenya 
Colony. 

c^ juv. d" (M. C. Z. 271034-5) Mihunga Ridge, U. 11 & 17. i. 39. 
Native names. Nyarukaka (Lutoro); erisibi (Lukonjo). 

Lanius collurio collurio Linnaeus 

Lanius collurio Linnaeus, 1758, Syst. Nat. ed. 10, 1, p. 94: Europe (Sweden, 
restricted by Hartert, 1907, Vog. pal. Fauna, 1, p. 439). 

9 (M. C. Z. 271036) Kitaya, T. T. 29. iii. 39. 
Native name. Namileu (Kimakonde). 

Laniarius ferrugineus major (Hartlaub) 
Telephonus major Hartlaub, 1848, Rev. Zool., p. 108: Elmina. Gold Coast. 
d" (M. C. Z. 271037) Mushongero, U. 1. ii. 39. 
Native name. Nsetcherezi (Lugezi). 

Laniarius ferrugineus mossambicus (Reichenow) 

Dryoscopus major mossambicus Reichenow, 1880, Journ. f. Orn., p. 141: 
Mozambique. 

9 (M. C. Z. 271038) Kitaya, T. T. 25. iii. 39. 
d^ 3 9 9 (M. C. Z. 271039-42) Mikindani, T. T. 17. iv. 39. 
d" (M. C. Z. 271043) Nchingidi, I. T. 11. v. 39. 

Native names. Lijaka (Kiyao); namgoii or nangoti (Kimakonde at 
Kitaya and Mikindani respectively). 

Laniarius luhderi luhderi (Reichenow) 
Dryoscopus liihderi Reichenow, 1874, Journ. f. Orn., p. 101: Cameroon. 
4 cTcf 1 9 (M. C. Z. 271044-7) Idjwi Id., B. C. 20. ii-1. iii. 39. 

Wings of d' d", 85.5, 86, 88 and 89.5 mm., of 9 , 80 mm. As the males 
have the measurements of the nominate form, the female is referred 
to it also. 

Native name. Ijongo (Lulegti). 



262 bulletin: museum of comparative zoology 

Dryoscopus cubla hamatus Hartlaub 

Dryoscopus hamatus Hartlaub, 1863, Proc. Zool. Soc. London, p. 106: Kazeh, 
i.e. Tabora, Unyamwezi, Tanganyika Territory. 

cf 9 (M. C. Z. 271048-9) Kitaya, T. T. 29-30. iii. 39. 
9 (M. C. Z. 271050) Mikindani, T. T. 18. iv. 39. 
d" (M. C. Z. 271051) Mbanja, T. T. 27. iv. 39. 
d" (M. C. Z. 271052) Lindi, T. T. 31. v. 39. 

Native navies. Lichau (Kiyao) ; mtiku (Kimakonde at Kitaya) ; 
lingoti (Kimakonde at Mikindani). 

Dryoscopus gambensis erwini Sassi 

Dryoscopus gambensis erwini Sassi, 1923, Orn. Monatsb., 31, p. 109: Forest 
west of Lake Tanganyika, Belgian Congo. 

c? (M. C. Z. 271053) Mihunga Ridge, U. 18. i. 39. 

Tschagra australis congener (Reichenow) 

Pomatorhynchus australis congener Reichenow, 1902, p. 258: (Type in Berlin 
Museum from Neu Helgoland, Songea district, Tanganyika Territory). 

d' (M. C. Z. 271054) Kitaya, T. T. 28. iii. 39. 
cf (M. C. Z. 271055) Mbanja, T. T. 3. v. 39. 
d (M. C. Z. 271056) Nchingidi, T. T. 10. v. 39. 

Native names. Kaviko (Kiyao); nanJcambawIa (Kimakonde). 

Tschagra australis littoralis (van Someren) 

Harpolestes australis littoralis van Someren, 1921, Bull. Brit. Orn. Club, 41, 
p. 102: Changamwe, Kenya Colony. 

d (M. C. Z. 271057) Magrotto Mtn., T. T. 7. vii. 39. 
Native name. Kishimdemlima (Kisambara). 

Tschagra senegala mozambica (van Someren) 

Harpolestes senegalus mozambicus van Someren, 1921, Bull. Brit. Orn. Club, 41, 
p. 103: Lumbo, Mozambique. 

d 9 (M. C. Z. 271058-9) Kitaya, T. T. 30. iii. 39. 
d (M. C. Z. 271060) Mbanja, T. T. 3. v. 39. 

These are somewhat intermediate between mozambica and orientalis 
(Cabanis), but a trifle nearer the former. 

Native names. Not differentiated from those of the last species. 



PETERS AND LOVERIDGE: AFRICAN BIRDS 263 

Antichromus anchietae reichenowi (Neumann) 

Telephonus reichenoivi Neumann, 1900, Journ. f. Orn., p. 120: Tanganyika 
Territory. 

d^ (M. C. Z. 271061) Magrotto Mtn., T. T. 5. vii. 39. 

This appears to be a perfectly good race as defined in Reichenow's 
synopsis of the species as evidenced by our reasonably good series of 
the three forms. 

Native name. Kishimdemlima (Kisambara, but not generic; see 
above). 

Nicator chloris chloris (Valenciennes) 
Lanitis chloris Valenciennes, 1826, Diet. Sci. Nat., 40, p. 226: Galam, Senegal. 
& cT (M. C. Z. 271062-3) Budongo Forest, U. 25. xi & 1. xii. 38. 



STURNIDAE 

CiNNYRiciNCLUs LEucoGASTER vERREAUXi (Finsch & Hartlaub) 

Pholidauges verreauxi "Bocage in litt.", 1870, Finsch & Hartlaub, Vogel Ost- 
Afr., p. 867: Caconda, Angola. 

c? juv., 9 (M. C. Z. 271064-5) Kitaya, T. T. 3. iv. 39. 

Native names. Litimbilwa (Kiyao); namitema (Kimakonde). 

Lamprocolius chalybeus sycobius Hartlaub 
Lamprocolius sycobius Hartlaub, 1859, Journ. f. Orn., p. 19: Tete, Mozambique. 
9 (M. C. Z. 271066) Mikindani, T. T. 14. iv. 39. 

Lamprocolius splendidus splendidus (Vieillot) 

Turdus splendidus Vieillot, 1822, Enc. Meth., 2, p. 653: Malimba, French 
Congo. 

9 9 (M. C. Z. 271067-8) Kibale Forest, U. 12. xii. 39. 

Onychognathus tenuirostris (Riippell) 

Lamprotornis tenuirostris Ruppell, 1836, Neue Wirbelth., Vogel, p. 26, pi. x, 
fig. 1 : Ethiopia. 

c^cf 9 (M. C. Z. 271069-71) Idjwi Id., B. C. 2. iii. 39. 

Native name. Kalungu (Lulega). 



264 bulletin: museum of comparative zoology 

NECTARINIIDAE 

Nectarinia kilimensis kilimensis Shelley 

Nectarinia kilimensis Shelley, 1884, Proc. Zool. Soc. London, p. 555: Mt. 
Kilimanjaro, circa 5000 feet, Tanganyika Territory. 

4 d^ cf 1 9 (M. C. Z. 271072-76) Mihunga Ridge, U. 18-19. i. 39. 
2cf d' 1 9 (M. C. Z. 271077-8) Idjwi Id., B. C. 24-27. ii & 1. iii. 39. 

Native names. Nswehe (Lukonjo) ; mononi (Lulega) ; msozi (Kis- 
ambara). In every case these names apply to all species of sunbirds. 

Nectarinia purpureiventris (Reichenow) 

Cinnyris purpureiventris Reichenow, 1893, Orn. Monatsb., 1, p. 61: Migere, S. 
of Lake Edward, Kivu district. 

c? d" (M. C. Z. 271079-80) Mihunga Ridge, U. 11 & 18. i. 39. 

Nectarinia erythrocerca erythrocerca Hartlaub 

Nectarinia erythrocerca Hartlaub, 1857, Syst. Orn. Westafr., p. 270: No 
locality (White Nile, S. of 8° N.,fide Heuglin, 1856). 

cf (M. C. Z. 271081) Mushongero, U. 3. ii. 39. 

Cinnyris cupreus cupreus (Shaw) 

Certhia cuprea Shaw, 1811, Gen. Zool., 8, p. 201: Malimba, French Congo. 

d^ (M. C. Z. 271082) Budongo Forest, U. 1. xii. 38. 
& (M. C. Z. 271083) Idjwi Id., B. C. 3. iii. 39. 

Cinnyris bifasciatus microrhynchus Shelley 

Cinnyris microrhynchus Shelley, 1876, Monogr. Nectarin., p. 219, pi. Ixvii: 
(Type in Brit. Mus. from Dar es Salaam, T. T.). 

d^ (M. C. Z. 271084) Lindi, T. T. 31. v. 39. 

d' (M. C. Z. 271085) Amboni Estate, T. T. 24. vi. 39. 

Cinnyris venustus igneiventris Reichenow 

Cinnyris igneiventris Reichenow, 1899, Orn. Monatsb., 7, p. 171: Karagwe, 
Uganda. 

d' (M. C. Z. 271086) Mubuku Valley, U. 7. i. 39. 
4 d^ d^ (M. C. Z. 271087-90) Mihunga Ridge, U. 12-18. i. 39. 
2 cf d^ (M. C. Z. 271091) Idjwi Id., B. C. 17-23. ii. 39. 



PETERS AND LOVERIDGE: AFRICAN BIRDS 265 

CiNNYRIS REICHENOWI REICHENOWI Sharpe 

Cinnyris reichenowi Sharpe, 1891, Ibis, p. 444: Sotik, Kenya Colony. 

5 d'c^ 3 9 9 (M. C. Z. 271092-99) Mihunga Ridge, U. 11-19. i. 39. 
8 cTd' 2 9 9 (M. C. Z. 271100-107) Idjwi Id., B. C. 18. ii-1. iii. 39. 



Cinnyris chloropygius orphogaster Reichenow 

Cinnyris chloropygius orphogaster Reichenow, 1899, Orn. Monatsb., 7, p. 169: 
Central Africa (Type in Berlin Mus. from Bukoba, Tanganyika Terri- 
tory). 

4 cf cf 1 9 (M. C. Z. 271108-12) Budongo Forest, U. 24. xi-6. xii. 38. 

Wings of cTcf, 51, 52, 53, and an immature cT, 47 mm.; of 9, 
53 mm. 



Chalcomitra angolensis angolensis (Lesson) 

Cinnyris angolensis Lesson, 1830, Traite d'Orn., p. 295: coast of Angola 
(Malimba, French Congo, fide Shelley, 1900, Birds Afr., 2, p. 112). 

d^ (M. C. Z. 271113) Mabira Forest, U. 8. xi. 38. 



Chalcomitra senegalensis aequatorialis (Reichenow) 

Cinnyris aequatorialis Reichenow, 1899, Orn. Monatsb., 7, p. 171: Bukoba, 
Tanganyika Territory. 

9 (M. C. Z. 271114) Mabira Forest, U. 8. .xi. 38. 

Native name. Kanunansnhi (Luganda). 

Breeding. Netted on nest, which, with the eggs is preserved. The 
nest was attached to the end of a twig of a rose-apple tree {Egernia 
jambos — Myrtaceae) about ten feet from the ground. The nest is 
constructed of vegetable fibres, stripped bark, grass and a little lichen, 
lined internally with vegetable floss; it measures 200 mm. in length, 
by about 60 mm. in breadth. The clutch consisted of two fresh eggs, of 
which the ground color of one was whitish, of the other pale green, both 
streaked with dark purplish brown spots with smudged outlines, 
measuring 20.5 x 13 mm., and 21 x 13 mm. respectively. 



266 bulletin: museum of comparative zoology 



Chalcomitra senegalensis inaestimata (Hartert) 

Cinnyris guUuralis inaestimata Hartert, 1899, in Ansorge, Under the African 
Sun, App., p. 351: East Africa (Type in the ximerican Mus. from Dar es 
Salaam) . 

4 d^cT 2 9 9 (M. C. Z. 271115-20) Mikindani, T. T. 10-21. iv. 39. 
cf 2 9 9 (M. C. Z. 271121-23) Mbanja, T. T. 29. iv-2. v. 39. 
cf 9 (M. C. Z. 271124-25) Lindi, T. T. 31. v. 39. 

cf (M. C. Z. 271126) Amboni Estate, T. T. 24. vi. 39. 

Native names. Chiteri (Kimakonde) ; msozi (Kisambara). 



Cyanomitra verticalis viridisplendens (Reichenow) 

Cinnyris viridisplendens Reichenow, 1892, Journ. f. Orn., pp. 54, 132: Bukoba, 
Tanganyika Territory. 

7 d'c? 1 9 (M. C. Z. 271127-34) Mihunga Ridge, U. 11-19. i. 39. 
6 cf d^ 1 9 (M. C. Z. 271135-41) Idjwi Id., B. C. 17. ii-2. iii. 39. 



Cyanomitra olivacea neglecta Neumann 

Cyanomitra obscura neglecta Neumann, 1900, Journ. f. Orn., p. 297: ffibwezi, 
Ukamba, Kenya Colony. 

9 (M. C. Z. 271142) Nchingidi, T. T. 10. v. 39. 
cf cf 9 (M. C. Z. 271143-45) Magrotto Mtn., T. T. 29. vi & 13. vii. 
39. 

Anthreptes collaris zambesiana (Shelley) 

Anthodiaeta zambesiana Shelley, 1880, Monogr. Nectarin., p. 243, pi. iii: 
Shupanga, Zambezi River, Mozambique. 

cf & (M. C. Z. 271146-7) Kitaya, T. T. 3-4. iv. 39. 
cf (M. C. Z. 271148) Mikindani, T. T. 17. iv. 39. 

Native names. Kasaragwe (Kiyao); kitwi (Kimakonde). 



Anthreptes collaris elachior Mearns 

Anthreptes collaris elachior Mearns, 1910, Smiths. Misc. Coll., 56, No. 14, 
p. 5: Changamwe, Kenya Colony. 

c? 9 9 (M. C. Z. 271152-54) Magrotto Mtn., T. T. 29. vi & 10. 
vii. 39. 



PETERS AND LOVERIDGE: AFRICAN BIRDS 267 

Anthreptes collaris ugandae van Someren 

Anthreptes collaris ugandae van Someren, 1921, Bull. Brit. Orn. Club, 41, p. 113: 
Maraquet, i.e. Marakwet, Kerio Frovince, Kenya Colony. 
cf (M. C. Z. 271155) Budongo Forest, U. 29. xi. 38. 
9 9 (M. C. Z. 271156-7) Idjwi Id., B. C. 23-24. iii. 39. 

Anthreptes tephrolaema tephrolaema (Jardine & Fraser) 

Nectarinia iephrolaemus Jardine & Fraser, 1851, Contrib. Orn., p. 154: Fer- 
nando Fo. 

cf (M. C. Z. 271158) Budongo Forest, U. 3. xii. 38. 

ZOSTEROPIDAE 
Zosterops virens scotti Neumann 

Zosterops scotti Neumann, 1899, Orn. Monatsb., 7, p. 24: Yerua Forest, 

Euwenzori Mtns., 8000 feet. 
Zosterops schubotzi Eeichenow, 1908, Orn. Monatsb., 16, p. 160: Lower slopes 

of Ronssoro, i.e. western slopes Euwenzori Mtns., Belgian C ongo. 

9 (M. C. Z. 271159) Mubuku Valley, U. 7. i. 39. 
d'd' 9 {M. C. Z. 271160-2) Mihunga Fidge, U. 10-16. i. 39. 
3 9 9 (M. C. Z. 271163-4) Idjwi Id., B. C. 18-20. ii-2. iii. 39. 

Native V antes. Kaholiamaisu (Lukonjo); mesu (Lulega). 

Zosterops virens stuhlmanni Reichenow 

Zosterops stuhlmanni Eeichenow, 1892, Journ. f. Cm., p. 54: Bukoba, Tan- 
ganyika Territory. 

cf (M. C. Z. 271165) Mabira Forest, U. 14. xi. 38. 

This breeding male has been compared with a topotype of stuhl- 
manni from Chantwara, Bukoba. 

PLOCEIDAE 

Passer griseus mosambicus van Someren 

Passer griseus mosambicus van Someren, 1921, Bull. Brit. Orn. Club, 41, p. 114: 
Lumbo, Mozambique. 

9 (M. C. Z. 271166) Kitaya, T. T. 27. iii. 39. 
d'd' (M. C. Z. 271167-68) Mikindani, T. T. 21. iv. 39. 
9 (M. C. Z. 271169) Mbanja, T. T. 28. iv. 39. 
cT juv., 9 9 (M. C. Z. 271170-72) Lindi, T. T. 31. v. 39. 

Native names. Kiswesu (Kiyao); nahoma (Kimakonde). 



268 bulletin: museum of comparative zoology 



Passer griseus ugandae Reichenow 

Passer diffusus ugandae Reichenow, 1904, Vogel Afr., 3, p. 231: Uganda. 
9 (M. C. Z. 271173) Nyakabande, U. 26. i. 39. 



Petronia superciliaris (Blyth) 

Gymnorhis superciliaris Blyth, 1845, Journ. Asiatic Soc. Bengal, 14, p. 553: 
South Africa. 

cf (M. C. Z. 271174) Mbanja, T. T. 1. v. 39. 



Ploceus bicolor stictifrons (Fischer & Reichenow) 

Symplectes stictifrons Fischer & Reichenow, 1885, Journ. f. Orn., p. 373: Lindi, 
Tanganjdka Territory. 

9 (M. C. Z. 271175) Nchingidi, T. T. 12. v. 39. 



Ploceus bicolor kersteni (Finsch & Hartlaub) 

Sycobrotus kersteni Finsch & Hartlaub, 1870, Vogel Ost-Afr., p. 404, pi. vi: 
Zanzibar. 

cf (M. C. Z. 271176) Magrotto Mtn., T. T. 8. vii. 39. 



Ploceus stuhlmanni stuhlmanni (Reichenow) 

Symplectes stuhlmanni Reichenow, 1893, Orn. Monatsb., 1, p. 29: (Type in 
Berlin Mus. from Bukoba, Tanganyika Territory). 

o^ d" (M. C. Z. 271177-78) Mihunga Ridge, U. 11 & 14. i. 39. 
cT (M. C. Z. 271179) Idjwi Id., B. C. 3. iii. 39. 

Native names. Kisajida (Lutoro and Lukonjo); chundabi (Lulega). 



Ploceus capitalis dimidiatus (Antinori & Salvadori) 

Hyphantornis dimidiata Antinori & Salvadori, 1873, Atti R. Accad. Torino, 8, 
p. 360: Kasala, Anglo-Egyptian Sudan. 

d" (M. C. Z. 271180) Budongo Forest, U. 1. xii. 38. 



PETERS AND LOVERIDGE: AFRICAN BIRDS 269 

Ploceus nigriceps nigriceps (Layard) 

Hyphantornis nigriceps Layard, 1867, Birds S. Afr., ed. 1, p. 180: Kuruman, 
British Bechuanaland. 

8 c?c? (M. C. Z. 271181-88) Mikindani, T. T. 15. iv. 39. 
d^ (M. C. Z. 271189) Mbanja, T. T. 28. iv. 39. 

Native name. Vanyehe (Kimakonde). 

Ploceus nigriceps graueri Hartert 

Ploceus graueri Hartert, 1911, Bull. Brit. Orn. Club, 29, p. 21: Usumbura, 
Belgian Ruanda-Urundi. 

d' (M. C. Z. 271190) Nyakabande, U. 28. i. 39. 

This all black-headed weaver was in the same breeding colony, in 
the rest-camp grounds, as the very similar feminina, listed below, 
which, however, has the occipital region buffy orange. 

Ploceus cucullatus feminina (Ogilvie-Grant) 

Hyphantornis feminina Ogilvie-Grant, 1907, Bull. Brit. Orn. Club, 21, p. 15: 
Mokia, Toro district, Uganda. 

cf cf 9 (M. C. Z. 271191-3) Budongo Forest, U. 6. xii. 38. 
4 cf d' (M. C. Z. 271194-7) Nyakabande, U. 28. i. & 7. ii. 39. 

Breeding. On February 7, a nest and eggs were collected, but pos- 
sibly are those of the last mentioned species occupying the same 
colony. 

Ploceus ocularius suahelicus Neumann 

Ploceus ocularius suahelicus Neumann, 1905, Journ. f. Orn., p. 339: Lewa, 
Usambara Mtns., Tanganyika Territory. 

cf (M. C. Z. 271198) Kitaya, T. T. 28. iii. 39. 
d" 9 (M. C. Z. 271199-200) Magrotto Mtn., T. T. 13. vii. 39. 

Ploceus aureoflavus aureoflavus Smith 

Ploceus aureoflavus A. Smith, 1839, 111. Zool. S. Afr., Aves, pi. xxx, fig. 1: 
"W. Africa," errore; probably Zanzibar. 

3 cf c? (M. C. Z. 271201-3) Kitaya, T. T. 27 & 31. iii. 39. 
d 9 (M. C. Z. 271204-5) Mikindani, T. T. 15. iv. 39. 
9 (M. C. Z. 271206) Mbanja, T. T. 28. iv. 39. 

Native names. Njesi (Kiyao) ; lichendi (Kimakonde at Kitaya) ; 
vanyehe (Kimakonde at Mikindani, but not specific). 



270 bulletin: museum of comparative zoology 

Ploceus xanthops camburni (Sharpe) 

Hyphantornis camburni Sharpe, 1900, Bull. Brit. Orn. Club, 10, p. 35: Mt. 
Kenya, Kenya Colony. 

9 (M. C. Z. 271207) Idjwi Id., B. C. 22. ii. 39. 

Native name. KishoJcoshoko (Lulega). 

EuPLECTES nigroventris Cassin 

Euplectes nigroventris Cassin, 1848, Proc. Acad. Nat. Sci. Philadelphia, p. 66: 
Zanzibar. 

cf (M. C. Z. 271208) Kitaya, T. T. 25. iii. 39. 
cfcf 9 (M. C. Z. 271209-ll)Mikindani, T. T. 19. iv. 39. 

Native names. Chiunga (Kiyao); nanchilanga (Kimakonde). 

Euplectes hordacea changamwensis (Mearns) 

Pyromelana flammiceps changamwensis Mearns, 1913, Smiths. Misc. Coll., 61, 
no. 11, p. 5: Changamwe, near Mombasa, Kenya Colony. 

3 cf cf (M. C. Z. 271212-4) Mikindani, T. T. 10. iv. 39. 
9 (M. C. Z. 271215) Mbanja, T. T. 29. iv. 39. 
cf 9 (M. C. Z. 271216-7) Lindi, T. T. 2. vi. 39. 

cT (M. C. Z. 271218) Amboni Estate, T. T. 24. vi. 39. 

Native name. Gong we (Kisambara). 

Euplectes capensis zambesiensis (Roberts) 

Xanthomelas xanthomelas zambesiensis Eoberts, 1922, Ann. Transvaal Mus., 
8, p. 266: Villa Fereira, Boror, Mozambique. 

cf (M. C. Z. 271219) Kitaya, T. T. 28. iii. 39. 
d'd' (M. C. Z. 271220-21) Mikindani, T. T. 10. iv. 39. 

Wings of c^cf , 66.5, 66.7, 68 mm.; tails of o"c^, 14.9, 15.8 and 15.8 
mm. 

Native names. Chiunga (Kiyao) ; lichende (Kimakonde at Kitaya) ; 
nayichilanga (Kimakonde at Mikindani for all Bishop birds). 

Euplectes capensis xanthomelas Riippell 

Euplectes xanthomelas Ruppell, 1840, Neue Wirbelth. Vogel, p. 94: Temben and 
Simien, Ethiopia. 

9 (M. C. Z. 271222) Mabira Forest, U. 11. xi. 38. 
c? (M. C. Z. 271223) Idjwi Id., B. C. 21. ii. 39. 



PETERS AND LOVERIDGE : AFRICAN BIRDS 271 

The latter bird would be referable to E. c. sahinjo Reichenow (1910, 
Orn. Monatsb., 18, p. 161: Sabinio Volcano, Kivu) were that form 
recognizable. Wings of cf d^ from Addis Ababa are 77, 76, 74 mm.; 
tails 13, — , 13.5 mm., which may be contrasted with cf cf from Kilosa 
of 68.7, 67.5, 65 mm.; and tails of 16, 15.4, 13.7 mm. 

Native name. Kifigifigi (Lulega). 

Spermestes cucullatus cucullatus Swainson 

Spermestes cucullatus Swainson, 1837, Birds W. Afr., 1, p. 201: Senegal. 

? (M. C. Z. 271224) Mabira Forest, U. 8. xi. 38. 
cf (M. C. Z. 271225) Budongo Forest, U. 24. xi. 38. 

Spermestes cucullatus scutatus Heuglin 

Spermestes scutatus Heuglin, 1863, Journ. f. Orn., p. 18: Dembea, Ethiopia. 

cf 9 (M. C. Z. 271226-7) Kitaya, T. T. 25. iii. 39. 
d" (M. C. Z. 271228) Mikindani, T. T. 19. iv. 39. 
9 (M. C. Z. 271229) Mbanja, T. T. 28. iv. 39. 

Native names. Kapingo (Kiyao) ; ujnngo (Kimakonde at Kitaya) ; 
utiri (Kimakonde at Mikindani). 

Spermestes bicolor stigmatophorus Reichenow 

Spermestes stigmatophorus Reichenow, 1892, Journ. f. Orn., p. 46: Bukoba, 
Tanganyika Territory. 

c^ (M. C. Z. 271230) Idjwi Id., B. C. 28. ii. 39. 

Native name. Ganger a (Lulega). 

Spermestes nigriceps nigriceps Cassin 

Spermestes nigriceps Cassin, 1852, Proc. Acad. Nat. Sci. Philadelphia, p. 185: 
Zanzibar. 

d^ (M. C. Z. 271231) Magrotto Mtn., T. T. 6. vii. 39. 
Native name. Mtongo (Kisambara). 

NiGRiTA canicapilla schistacea Sharpe 
Nigrita schistacea Sharpe, 1891, Ibis, p. 118: Sotik, Kenya Colony. 
9 (M. C. Z. 271232) Budongo Forest, U. 1. xii. 38. 



272 bulletin: museum of comparative zoology 



Spermophaga ruficapilla ruficapilla (Shelley) 

Spermospiza ruficapilla Shelley, 1888, Proc. Zool. Soc. London, p. 30: Bellima, 
Uelle district, Belgian Congo. 

9 (M. C. Z. 271233) Budongo Forest, U. 30. xi. 38. 

Enemies. From the .stomach of a bird snake (Thelotornis k. kirt- 
landii), killed at Buta, Uelle, B. C, we recently removed two large 
nestlings which Dr. J. P. Chapin has been able to identify generically 
by means of the three black palatal spots. Dr. Chapin states that 
ruficapilla has been collected at Buta by Hutsebaut, but the possibility 
remains that the rarer j^oliogenys may turn up there. 



Hypargos niveoguttatus (Peters) 

Spermophaga niveoguttaki W. Peters, 1868, Journ. f. Orn., p. 133: Inhambane, 
Mozambique. 

cf (M. C. Z. 271234) Mikindani, T. T. 17. iv. 39. 
Native name. Linguruguru (Kimakonde). 



Pytilm melba grotei Reichenow 

Pytelia [sic] melba grotei Reichenow, 1919, Journ. f. Orn., p. 227: Kionga, S, 
bank Rovuma River, Mozambique. 

cT d' (M. C. Z. 271235-6) Kitaya, T. T. 28. iii. 39. 

Almost topotypic, Kitaya being on the north bank just a few miles 
above Kionga. 

Native names. Kapalanganga (Kiyao); chanika (Kimakonde). 



Lagonosticta rubricata haematocephala Neumann 

Lagonosticta rubricata haematocephala Neumann, 1907, Orn. Monatsb., 15, 
p. 168: Songea, Tanganyika Territory. 

d" 9 (M. C. Z. 271237-38) Kitaya, T. T. 3. iv. 39. 
9 (M. C. Z. 271239) Mikindani, T. T. 21. iv. 39. 

Native names. Kapelepete (Kiyao); ckinonombc (Kimakonde at 
Kitaya); chididi (Kimakonde at Mikindani). 



PETERS AND LOVERIDGE: AFRICAN BIRDS 273 



CoccoPYGiA MELANOTis NYANZAE (Neumann) 

Neisna dufresneyi nyansae Neumann, 1905, Journ. f. Orn., p. 350: Bukoba, 

Tanganyika Territory. 
Neisna minima Ogilvie-Grant, 1906, Bull. Brit. Orn. Club, 16, p. 117: Mubuku 

Valley, Ruwenzori Mtns., 6000 feet, Uganda. 

3 d"o" 2 9 9 (M. C. Z. 271240-4) Mubuku Valley, U. 2 & 7. i. 39. 
cf 9 (M. C. Z. 271245-6) Idjwi Id., B. C. 27. ii. 39. 



The Ruwenzori series are topotypes of minima. 
Native name. Katwasundi (Lukonjo). 



EsTRiLDA ASTRILD MINOR (Cabanis) 

Habropyga minor Cabanis, 1878, Journ. f. Orn., p. 229: Voi River, Kenya 
Colony. 

c^ (M. C. Z. 271247) Nchingidi, T. T. 17. v. 39. 
9 (M. C. Z. 271248) Siga Caves, T. T. 8. vi. 39. 
2 c? cf 2 9 9 (M. C. Z. 271249-52) Magrotto Mtn., T. T. 29. vi. & 
5. vii. 39. 



Native name. Mseke (Kisambara). 



EsTRiLDA ASTRILD NYANZAE Neumann 

Estrilda astrild nyanzae Neumann, 1907, Journ. f. Orn., p. 596: Bukoba, 
Tanganyika Territory. 

d^ 9 (M. C. Z. 271253-4) Idjwi Id., B. C. 21. ii & 2. iii. 39. 

Native name. Roboneka (Lulega). 



Estrilda nonnula nonnula Hartlaub 

Astrilda [sic] nonnula Hartlaub, 1883, Journ. f. Orn., p. 425: Kudurma, Bahr el 
Ghazal, Anglo-Egyptian Sudan. 

3 9 9 (M. C. Z. 271255-57) Budongo Forest, U. 28. xi-6. xii. 38. 
1 c^ 3 9 9 (M. C. Z. 271258-61) Mihunga Ridge, U. 12. i. 39. 
3 c?cf 3 9 9 (M. C. Z. 271262-5) Idjwi Id., B. C. 18. ii-3. iii. 39. 



Native name. Fnnzi (Lulega). 



274 bulletin: museum of comparative zoology 



Uraeginthus angolensis nlassensis Reichenow 

Uraeginthus bengalus niassensis Reichenow, 1911, Mitt. Zool. Mus. Berlin, 6, 
p. 228: Songea, Tanganyika Territory. 

d" 9 (M. C. Z. 271266-7) Kitaya, T. T. 31. iii. 39. 
c? 9 9 (M. C. Z. 271268-70) Mikindani, T. T. 18-19. iv. 39. 
3 d'd' S 9 9 (M. C. Z. 271271-6) Mbanja, T. T. 26. iv. 39. 

Native names. Kitwittvitwi (Kiyao) ; chididi (Kimakonde at Kitaya 
and Mikindani, but not specific). 



Vidua macroura (Pallas) 

Fringilla macroura Pallas, 1764, in Vroeg, Cat., Adumbrat., no. 144, p. 3: "East 
Indies," (Angola, ^de Edwards & Brisson). 

9 (M. C. Z. 271277) Budongo Forest, U. 1. xii. 38. 
d^ (M. C. Z. 271278) Idjwi Id., B. C. 2. iii. 39. 
c? (M. C. Z. 271279) Kitaya, T. T. 30. iii. 39. 
cf (M. C. Z. 271280) Mikindani, T. T. 18. iv. 39. 
(f (M. C. Z. 271281) Mbanja, T. T. 6. v. 39. 

All the males, with the exception of the Kitaya bird, possess full 
nuptial tail feathers. 

Native names. Lidangiragira (Lulega) ; chiunga (Kiyao) ; nandumbi 
(Kimakonde at Kitaya and Mikindani). 



Steganura paradisaea (Linnaeus) 

Emberiza paradisaea Linnaeus, 1758, Syst. Nat. ed. 10, 1, p. 178: Africa (re- 
stricted to Angola in ed. 12, p. 312). 

cT (M. C. Z. 271282) Mbanja, T. T. 6. v. 39. 

In full breeding plumes. We follow Chapin (1922, Amer. Mus. Nov., 
no. 43, pp. 1-12) who, with more material, decided to separate 
paradisaea from the aucupum group on account of the overlap in 
ranges, and as both species have been recorded as breeding in the same 
localities. 



PETERS AND LOVERIDGE: AFRICAN BIRDS 275 

FRINGILLIDAE 

POLIOSPIZA STRIOLATA GRAUERI (Hartett) 

Serinus striolata graueri Hartert, 1907, Bull. Brit. Orn. Club, 19, p. 84: Ruwen- 
zori Mtns. 7000 feet, Uganda. 

9 (M. C. Z. 271283) Mihunga Ridge, U. 16. i. 39. 
Native name. Kiswcre (Lukonjo). 

PoLiosPizA BURTONi TANGANJICAE (Granvik) 

Serin-US alMfrons tanganjicae Granvik, 1923, Journ. f. Om., Sonderheft, p. 191: 
Lake Tanganyika (ex. Reichenow MS.). 

d^ 9 (M. C. Z. 271284-5) Mihunga Ridge, U. 10. i. 39. 
Native 7iame. Kiswere (Lukonjo). 

Spinus citrinelloides frontalis Reichenow 

Spinus citrinelloides frontalis Reichenow, 1904, Vogel Afr., 3, p. 275: (Type in 
Berlin Mus. from Lake Kivu). 

d" 9 (M. C. Z. 271286-7) Mihunga Ridge, U. 10 & 16. i. 39. 
9 (M. C. Z. 271288) Idjwi Id., L. Ki\Ti, B. C. 28. ii. 39. 

Native names. Kasande or Kanyangaleka (Lukonjo); kararamazi 
(Lulega). 



EXPLANATION OF PLATES 



PLATE 1 



Peters and Loveridge — African Birds. 



PLATE 1 
Map showing Principal Collecting Localities 

1938 

Landing at Mombasa (25.x), except for a stopover at Naivasha and Kin- 
angop (26-3Lx), Loveridge proceeded by rail direct to Jinja (1-5. xi). Thence 
to Mabira Forest (5-21.xi), Budongo Forest (22.xi-7.xii), Kibale Forest 
(8-19. xii), Bundibugyo near Bwamba Forest (19-26. xii), Bugoye, foot of 
Ruwenzori Mountains (26-28.xii) and Mubuku Valley at 7000 ft. (29.xii-). 

1939 

On leaving Mubuku (9.i) Loveridge descended down the valley to Mihunga, 
circa 6000 ft. (9-21. i), then back to Bugoye (21-24.i), Nyakabande (25-30.i), 
Mushongero (30.i^.xi), returned to Nyakabande (4-8.ii); Kisenyi (8-13.ii), 
Goma (13-14. ii), Mamvu on Idjwi Island (14-16.ii), Upper Mulinga on Idjwi 
(16.ii-6.iii), Uvira (7-8.iii), Ujiji (9-16. iii), Dar es Salaam (18-19.iii), Mikin- 
dani (22-24.iii), Kitaya (24.iii-7.iv), Mikindani (7-24.iv), Mbanja (25.iv- 
6.v), Lake Rutamba (6-8.v), Nchingidi (9-21. v), Lindi (22.v-4.vi), Siga 
Caves (7-17, vi). Amboni Estate (17-27. vi), Magrotto Mountain (27.vi- 
21.vii), Tanga (21-23.vii), Ivilindini (24-26. vii). 



BULL. MUS. COMP. ZOOL. 



Peters and Loveridge, African Birds. Plate 1. 



MuhuKu Valley 




'Bud on (^0 
^ Forest 

•.--.Kihale Fo'-est^ 
-^ / , JINJ.'^ 

'V^^- Forest^ 






CosUrmansville ■< 
Uvira 



BELGIAN 
CONGO 




-^-.-v C 9 



PLATE 2 



Petebs and Loveridge — African Birds. 



PLATE 2 

Fig. 1. Nestling Crowned Crane {Balearica p. gibbericeps) 

The beautiful crowned crane is protected throughout East Africa, despite 
the fact that in certain locahties it destroys seed-corn. The natives of Ruanda 
appear to be unaware of the regulations as several nestlings were offered for 
sale at Nyakabande, and a clutch of eggs at Mushongero, on Lake Mutanda. 

Fig. 2. Nest of White-breasted Cormorant {Phalacrocorax c. lugubris) 

A huge colony of these birds had their nests on a little island in Lake Mu- 
tanda, but when we visited the nests on February 1, the young had all left 
them. An immature pair, which had been feeding on catfish, established 
identity, and seem to constitute the first record of the Red Sea race breeding 
in the Kivu region, according to Dr. J. P. Chapin. 



BULL. MUS. COMP. ZOOL. 



Peters AND LovERiDGE. African Birds. Plate 2. 





PLATE 3 



Peters and Lovebidge — African Birds. 



PLATE 3 

Fig. 1. Camp at edge of Budongo Forest, Uganda. 
The Budongo constitutes an extensive belt of primeval forest lying to the 
east of Lake Albert in northwestern Uganda. The drying safe in the foreground 
protects bird-skins from ovulating flies and other pests, furnishes shade, and 
facilitates their quick removal to cover in sudden downpours. The safe is so 
light it can be carried on a porter's head, or can be quickly taken down and 
packed flat for truck transport. 

Fig. 2. KizAMBA, A Mganda, skinning birds 

Both Baganda skinners rapidly mastered their work and turned out good 
skins. Wherever we went they were a source of wonderment and interest to 
the local youngsters. In the background can be seen the trees surrounding the 
Rest Camp at Nyakabande, in which flocks of noisy weavers were nesting. 
The colony was composed of two species, distinct, yet strangely similar. 



BULL. MUS. COMP. ZOOL. 



Peters AND LovERiDOE. African Birds. Plate3. 













Bulletin of the Museum of Comparative Zoology 

AT HARVARD COLLEGE 

Vol. LXXXIX, No. 6 



GROWTH AND DEVELOPMENT OF THE PROBOSCIS 

MONKEY 



By Adolph H. Schultz 

Laboratory of Physical Anthropology 
John Hopkins University 



With Four Plates 



CAMBRIDGE, MASS., U.S.A. 

PRINTED FOR THE MUSEUM 

March, 1942 



No. G. — Growth and Development of the Proboscis Monkey 
By Adolph H. Schultz 

INTRODUCTION 

In the study of taxonomic and phylogenetic problems relating to 
primates very little consideration has been given so far to the con- 
ditions of growth and development. The changes with age in an animal 
or man are more profound and extensive than the changes with geo- 
logical age in the evolutionary history of the species. The reconstruc- 
tion of the latter is largely dependent upon luck in finding fossil 
remains, limited at best to skeletal and dental parts, and will always 
have to rest hea\aly upon indirect evidence of a comparative-anatomi- 
cal nature. For the construction of complete ontogenetic histories the 
necessary materials can be made available, even though this is often 
far from easy. Species characters are in the last analysis nothing but 
the end results of inherited ontogenetic processes, and alterations in 
the latter bring about changes in the former. We possess surprisingly 
little reliable and comprehensive information regarding the similarities 
and dissimilarities in the ontogenetic conditions of primates. Thus we 
are still quite unable to state fully to what extent and in which re- 
spects the growth of man differs from the growth of apes and monkeys, 
yet such knowledge promises to shed new light on the great problem 
of the origin of man. 

The present paper is intended chiefly as a record of the conditions 
of growth and development in an Old World monkey, and as such is 
to be regarded as a contribution toward the eventual accumulation of 
data sufficient for general comparisons of the ontogenetic conditions 
m primates. The proboscis monkey {Nasalis larvatus), which is the 
subject of this study, belongs to the catarrhine subfamily of Semno- 
pithecinae, a group of primates for which there exists particularly scant 
information regarding growth. Breschet (1845) has pictured and 
briefly described two fetuses of the proboscis monkey and Keibel 
(1906) has contributed a few data on and illustrations of two other 
fetuses of the same primate species. Schwalbe (1911) has listed some 
measurements of four fetuses of the proboscis monkey and has given 
a detailed description of the direction of the hair in the oldest of these 
specimens. The measurements by Schwalbe and those by Breschet 



280 bulletin: museum of comparative zoology 

are of very limited value chiefly because these authors failed to define 
adequately their methods of measuring. Wiedersheim (1901a) re- 
ported on one fetus and one juvenile proboscis monkey, dealing 
principally with the anatomy of the nose. The author (1935), finally, 
has published a tentative account of the sequence of eruption of the 
permanent teeth in Nasalis and investigated the age changes in the 
relative length of the various regions of the spinal column (1938a) 
and the age changes in the proportionate size of the eye and the orbit 
(1940 b) in the same primate. As far as the writer can ascertain these 
are the only papers dealing with one or another phase of growth and 
development in the proboscis monkey. 

It must be mentioned here that the technique of the measurements 
and proportions used in this paper has already been described in full 
detail in a special publication by the author (1929) and, hence, can 
not be repeated here. 

This investigation represents one of the results of the Asiatic Pri- 
mate Expedition of 1937 in which the writer had participated. This 
expedition had been organized and aided in manifold ways by Mr. 
H. J. Coolidge, Jr. of the Museum of Comparative Zoology of Harvard 
University. Permission for the collecting of a series of proboscis 
monkeys was granted by the authorities of British North Borneo, 
who facilitated the work of the expedition in the most generous 
manner. The author is particularly indebted to Mr. H. G. Keith, 
Conservator of Forests of British North Borneo, for his many helpful 
acts which contributed enormously to the success of the expedition. 
During the field-work in Borneo the writer was very ably assisted by 
his friend. Dr. S. L. Washburn, whose conscientious and untiring 
labors greatly furthered the collection of the specimens and data used 
for this paper. 

Finally, the author wishes to express his sincere thanks to Prof. M. 
Brodel for the excellent portrait studies of adult proboscis monkeys 
which appear as plates 1, 2 and 3. These are based upon plaster death- 
masks which the writer had made in the field on freshly killed speci- 
mens. 

MATERIAL 

The series of proboscis monkeys, studied by the author, consist of a 
total of 51 bodies and one of 41 skeletons (and 2 skulls). The entire 
skeletal series is derived from the series of bodies. Among the latter 
there are 6 fetuses, 13 infants, 7 juveniles, 15 adult females, and 10 
adult males. The term "infant" is applied to all specimens with in- 



SCHULTZ: THE PROBOSCIS MONKEY 281 

complete or complete deciduous dentition, but as yet no permanent 
teeth. "Juveniles" are specimens with at least one, but not all per- 
manent teeth completely erupted. A specimen is classified as "adult" 
when its permanent dentition is complete regardless of the state of 
epiphyseal union or of cranial suture closure. An occasional reference 
to an "old" animal indicates that there exists an advanced degree of 
attrition of the teeth. The series of skeletons consists of 9 infants (in 
addition to which some X-ray photographs could be consulted), 7 
juveniles, and 25 adults. Forty-two of these specimens (including 
two fetuses) were collected on the Asiatic Primate Expedition of 1937 
at Abai on the Kinabatangan river in British North Borneo. All this 
material was weighed, measured and autopsied in the field, and all 
specimens of postnatal ages, except the youngest infant, were made 
into skeletons. All the skeletons, except two which had been given to 
the author, belong to the Museum of Comparative Zoology of Harvard 
University. The writer is very grateful to Dr. T. Barbour and Mr. 
H. J. Coolidge, Jr. for the loan of this material over a period of more 
than two years. Three fetal and one infantile proboscis monkeys have 
been measured by the author at the U. S. National Museum, with the 
kind permission of Mr. G. S. Miller, Jr., and one fetal and two in- 
fantile specimens had been examined by the writer at the Wistar In- 
stitute of Anatomy when under the directorship of the late Dr. M. J. 
Greenman. The body of one proboscis monkey from Dutch Borneo 
had generously been sent to the author by Dr. L. M. Huey of San 
Diego. This specimen was later made into a skeleton. In the author's 
collection there are also two skulls of adult Nasalis larvatus which 
come from Dutch Borneo and have been made use of in this study. 
The above-mentioned specimens from the National Museum and those 
from the Wistar Institute also come from various localities in Dutch 
Borneo. It may be mentioned here that in regard to the characters 
dealt with in this paper no differences could be detected between the 
proboscis monkeys from different parts of Borneo. 

GROWTH IN GENERAL 

Nothing is known in regard to the durations of the various periods 
of growth in the proboscis monkey, but it may be assumed that these 
do not differ radically from the conditions in macaques. In the latter 
pregnancy is known to last 166 days (Hartman, 1932) and the perma- 
nent dentition is completely erupted at the age of 7 years (Schultz, 
1940a). 



282 bulletin: museum of comparative zoology 

The youngest proboscis monkey infant collected showed a conspicu- 
ous and quite fresh umbilical scar and had only a few milk incisors just 
appearing above the gums. This specimen weighed 0.454 kg., or only 
little more than the oldest fetus. It can be safely concluded, therefore, 
that this infant had been born quite recently and that the weight of 
the proboscis monkey at birth equals approximately 0.45 kg. 

The weight of adult proboscis monkeys averages in females (15 
specimens) 9.873 kg. and in males (10 specimens) 20.344 kg. Indi- 
vidually the weight varies among these fully adult specimens between 
8.165 and 11.794 kg. in females and between 14.062 and 23.587 kg. in 
males. This sex difference in size in the proboscis monkey, showing a 
ratio of roughly 1:2, is equalled among simian primates by only a 
few forms, such as orang-utan, gorilla, and some baboons. 

The weight of the newborn in percentage of the average weight of 
adult females amounts to 4.6 in the proboscis monkey. The corre- 
sponding figures for other primates are, as far as could be established 
so far, 6.7 in macaques, 4.1 in orang-utans, 4.0 in chimpanzees, and 
5.5 in man (Schultz, 1941). 

In contrast to macaques and orang-utans, proboscis monkeys seem 
not to become pregnant until their permanent dentition is comple- 
ted. Out of 15 adult female proboscis monkeys only two were preg- 
nant, one containing a very small embryo and the other double- 
ovum twins of the second half of intrauterine development. Four 
other females, collected at practically the same time, had very large 
mammary glands containing a copious supply of fresh milk and were 
accompanied by infants of varying size. These data support the con- 
clusions that there can not exist any breeding season for proboscis 
monkeys and that their fertility is comparatively low^ . 

GROWTH CHANGES IN BODY PROPORTIONS 

The proportions of the outer body undergo changes with advancing 
age due to differing rates of growth in the bodily parts to which the 
proportions refer. The growth changes in the main proportions of the 
trunk of the proboscis monkey are shown by the data in table 1. The 
chest circumference (at level of insertion of fourth pair of ribs) in per- 
centage of the trunk height (from supra-sternal notch to upper end of 
pubic symphysis) decreases with growth, but increases again in males 
during adult life. The general ontogenetic drop in this index agrees 

lAmong 14 adult female Macaca irus, collected at the same place and time, 7 were pregnant, 
some having very small embryos, others nearly full-term fetuses. 



SCHULTZ: THE PROBOSCIS MONKEY 



283 



witli the findings in all other primates examined so far (Schultz, 1926, 
1937) and a marked, late, secondary rise in males occurs also in orang- 
utan (Schultz, 1941) and in man. Judging by samples of other adult 
lower catarrhines, among which this index varies between 83 and 129 
(Schultz, 1933a), the proboscis monkey is characterized by a com- 
paratively stout trunk, particularly the males, though even the latter 
have a much more slender trunk than any of the higher primates (see 
also table 5). 



Table 1 
Age changes in the proportions of the trunk and tail in proboscis 
monkeys. Averages (above) and ranges of variations (omitting deci- 
mals, below). 







Relative 


Relative 


Relative 




Relative 


Age 


Speci- 


Chest 


Shoulder 


Hip 


Chest 


Tail 




mens 


Circumfer. 


Breadth 


Breadth 


Index 


Length 


Fetuses 


6 


146.9 


46.4 


37.8 


95.2 


162.7 






137-165 


44-51 


35-40 


93-100 


143-192 


Infants 


13 


121.4 


39.8 


37.4 


96.3 


190.8 






112-137 


33-46 


35-40 


91-104 


166-209 


Juveniles 


7 


110.4 


33.4 


33.3 


96.6 


169.8 






99-120 


31-37 


31-35 


93-100 


162-184 


Adults 9 


15 


108.0 


33.0 


33.6 


99.5 


152.3 






96-120 


30-38 


32-35 


95-103 


139-169 


Adults d" 


10 


125.0 


38.0 


36.8 


99.6 


156.6 






112-135 


36-42 


34-40 


95-104 


143-167 



The percentage relation between the shoulder breadth and the 
trunk height decreases "with advancing age in the proboscis monkey, as 
in the other primates studied so far. Adult males of the proboscis 
monkey, as those of chimpanzee and of orang-utan (Schultz, 1941), 
acquire secondarily relatively broad shoulders in connection with the 
widening of the chest. 

The relative hip breadth (bitrochanteric diameter in percentage of 
trunk height) changes comparatively little during growth, though its 



284 bulletin: museum of comparative zoology 

general trend is to decrease somewhat. Adult males possess a higher 
average index than adult females. This, however, must be due to 
stronger heads and necks of the femora in the larger males, since the 
relative width of the pelvic inlet is slightly greater in females than in 
males. The latter proportion (pelvic inlet breadth in percentage of 
trunk height) averages 13.2 in adult females and 12.9 in adult males; 
in infants and in juveniles it averages only 10.7. A closely correspond- 
ing age change and sex difference in this index has been found also 
in chimpanzee and in orang-utan (Schultz, 1940a, 1941). Incidentally, 
the pelvic inlet is proportionately larger in females than in males also 
in its sagittal diameter, as shown by the following averages for adult 
proboscis monkeys: Promontorium to symphysion in percentage of 
trunk height = 18.3 in females and 16.4 in males. 

The chest index (transverse diameter of chest in percentage of 
sagittal diameter, at same level as chest girth) increases slightly with 
growth, but remains in its averages below 100, in contrast to all higher 
primates in which it rises to way above 100 during postnatal growth; 
i. e., the chest becomes much wider than it is deep. The shape of the 
thoracic cavit^^ undergoes also some change with age in the proboscis 
monkey, as shown by figure 1. The most noteworthy of these changes 
is the slight, but nevertheless significant shift of the spinal column 
toward the center of the thorax. This ontogenetic, topographic 
change is much more pronounced in man and the great apes than in the 
lower catarrhines (see figure 1 and, for further discussion, Schultz, 
1941). 

The relative length of the tail (tail length, measured with tape, 
ventrally, from root to tip of outer tail, in percentage of trunk height) 
undergoes very considerable changes with age in the proboscis monkey, 
as shown by the data in table 1. This proportion increases at first, to 
reach its maximum values during infantile life, and decreases subse- 
quently until cessation of growth in adult life. This decrease in relative 
tail length during postnatal growth is the general rule in tailed mon- 
keys, as has been demonstrated in other publications by the author 
(1938 and, by repeated measurements on living macaques, 1933b). 

The growth changes in the most significant proportions of the limbs 
of the proboscis monkey are shown in table 2. The average relative 
lower limb length (thigh length -f knee height in percentage of trunk 
height) increases during the latter part of fetal life, reaching its maxi- 
mum values among infants, and decreases somewhat thereafter. 
Closely corresponding age changes take place in the relative upper 
limb length (upper arm length -f- forearm length + hand length in per- 



SCHULTZ: THE PROBOSCIS MONKEY 



285 








« 




a 








"a 




Oi 




o 








'S 




eS 




t-, 




O 


60 


Xi 


_c 


-1-i 


'o 


o 


2 


-kJ 


-^ 


(H 


«4— 1 


^ 


o 






'a 


-1-3 


u 


^ 


-3 


M) 


a 


■<u 


(U 


^ 


a, 


<u 


<LI 


s 

03 


a 




02 


-u 




3 


o 


o 


-1-3 


^ 


-o 


.2? 


s 


'-3 


3 


> 


73 


CJ 








_o 


"S 


'S 


^ 


c3 


GQ 


^ 


X3 


O 


•t-H 


43 


;-! 


+j 


c«^ 




o 


13 
a; 


;-. 


-1^ 


S 


2 


o. 




^ 
t 


"> 


P 


o 


o 




«<-l 


U-i 


<M 


o 


o 


03 


a 


09 


a 
g 




.2 


-u 




tn 


c 


c3 


o 


"& 


c 


(4-^ 


_o 


o 


'-3 




t- 


CO 


Qi 


bC 


m 


_C 


C 






'G 


U-t 


c3 


O 


H 


% 




> 


T— ( 


<u 


bi) 








s 





286 



bulletin: museum of comparative zoology 



centage of trunk height). The ontogenetic alterations in these rela- 
tions between limb lengths and trunk height are at least in their direc- 
tion strikingly similar to the corresponding growth changes in macaque 
and in orang-utan (Schultz, 1937, 1941), but they differ from the con- 
ditions in man and chimpanzee (Schultz, 1926, 1940a). 



Table 2 

Age changes in the proportions of the limbs in proboscis monkeys. 
Averages (above) and ranges of variations (omitting decimals, be- 
low). 







Relative 


Relative 


Inter- 






Age 


Speci- 


Lower Limb 


Upper Limb 


membral 


C rural 


Brachial 




mens 


Length 


Length 


Index 


Index 


Index 


Fetuses 


6 


106.9 


141.9 


132.8 


85.2 


91.5 






97-122 


133-159 


129-137 


83-88 


87-96 


Infants 


13 


120.4 


152.2 


126.3 


87.4 


96.5 






112-129 


139-162 


121-153 


85-90 


91-100 


Juveniles 


7 


113.3 


137.5 


121.2 


88.7 


97.8 






109-116 


133-143 


117-124 


87-92 


93-102 


Adults 9 


15 


109.1 


132.2 


121.0 


87.7 


99.5 






105-115 


126-140 


119-124 


86-90 


94-102 


Adults cf 


10 


113.6 


137.6 


121.2 


88.1 


100.3 






108-118 


129-142 


119-125 


84-91 


98-102 



The intermembral index (total length of upper limb in percentage 
of total length of lower limb) decreases somewhat with advance in age, 
i. e., the lower limb grows faster than the upper one. The same gen- 
eral growth change in this proportion occurs in the macaque, chimpan- 
zee and man (Schultz, 1926, 1937, 1940a), but not in the orang-utan 
in whom this index is practically constant (Schultz, 1941). The crural 
index (length of leg in percentage of length of thigh) increases slightly 
with advancing growth and the brachial index (length of forearm in 



SCHULTZ: THE PROBOSCIS MONKEY 



287 



percentage of length of upper arm) rises considerably. Particularly 
an ontogenetic increase in the latter index represents the rule among 
primates, including man. 



\- 

X 

n]<o.o 

§9.0 



UJ 

ci:7.0 



-6.0 

>- 

^5.0 
< 



< 



2.0 
10 



* 




















4? 


























X 


i 


i 




^ 


^^'-- 


.-'--' 


f 


.V 


4 






d 




i 






^ 
y 

i."' 




(. 

t 




v~ 




absolute: capacity 




I 


V^ 


f 




f« 


f 






REL 


NA5ALI5 = f •* 
ATIVE CAPACITY : 


/ 




i. 


\ 











NA5ALI5 = ■ 
LANQUR = O 
MACAaUE = H , 


« 


ii 
















CHIMPANZEE = <3> 
ORANQ-UTAN = ^ 

1 






> 




. 


^oo 




























o 


o 
















1 
















O 
















■'-* 


B 
























^1 


^^a 


■■— . 








•--*t 


■0 — 


-- 






» 





^-10 



m 



90 



80 



TO 



6oS 



< 



40. 



30 i 



20 



10 



-H 13 'IS M 

BODY WEIGHT 



-19 21 23 k3. 



Fig. 2. Correlation between absolute and relative cranial capacity and body 
weight in the proboscis monkey and (for relative capacity only) in some 
other primates. 



The percentage relations between breadth and length of the foot 
and of the hand of proboscis monkeys are shown in table 3. Both the 
foot and the hand become more slender during growth, as they do in all 
the primates studied so far. The only exception to this general rule 
consists in the trend among males to broaden hand and foot again 
slightly late in growth. The relative length of the thumb (total thumb 
length in percentage of greatest total hand length) decreases with ad- 
vancing age in the proboscis monkey, as it does in most other primates 



288 



bulletin: museum of comparative zoology 



for which relevant data have become available. In all 51 specimens of 
Nasalis the fourth digits greatly surpass the second in length in the 
hand as well as in the foot and this regardless of age. In five instances 
the fourth fingers even equal the middle fingers in length. The toes 

Table 3 

Age changes in the proportions of the foot and the hand in proboscis 
monkeys. Averages (above) and ranges of variations (omitting 
decimals, below). « 







Relative 


Relative 


Relative 


Age 


Specimens 


Foot 


Hand 


Thumb 






Breadth 


Breadth 


Length 


Fetuses 


6 


26.4 


40.3 


49.2 






25-29 


37-42 


,48-51 


Infants 


13 


23.1 


33.5 


47.7 






21-25 


29-38 


42-51 


Juveniles 


7 


21.3 


28.9 


47.1 






20-22 


27-31 


44-51 


Adults 9 


15 


21.9 


28.5 


45.0 






20-23 


27-31 


42-48 


Adults cf 


10 


23.1 


29.0 


44.3 






22-24 


27-30 


41-50 



II to V are always webbed in the region of the basal phalanges. Be- 
tween toes II and III this webbing extends frequently as far as the 
middle of middle phalanx II, and in three cases it reaches even the 
distal end of the latter phalanx, or fully as far as in any specimens of 
Sym/phalangus syndactylus (Schultz, 1933 c, table 18). 

The growth changes in the most significant proportions of the head 
of the proboscis monkey are shown by the data in table 4 and are illus- 
trated in figure 3 and plates 1-3. Very few indices change during 
growth as extensively as the relative head size (arithmetic mean of 
head length, breadth and height in percentage of trunk height). On 
an average this proportion decreases in proboscis monkeys from 49 in 



SCHULTZ: THE PROBOSCIS MONKEY 



289 



art-er KeibelJ306 



5.H. = 25 




mm. 



A.PE. 414 



5.H. = 'l'I9rr. 



m. 




NahMus.^42223 



5.H. = 'I54-mm. 



Fig. 3. Side views of heads of proboscis monkey fetuses, posed in ear-eye 
horizon. S. H.= sitting-height. 



290 bulletin: museum of comparative zoology 

fetuses to only 17 in adults. The average head diameter gives only 
an approximate indication of the size of the brain and is more closely 
correlated with the latter in early than in late stages of growth. To 
gain some information on the growth of the brain, the cranial capacity 
has been measured (with rape seed) in the entire series of N^asalis 
skulls. This capacity averages 70.0 cc. in 9 infants, 83.1 cc. in 7 
juveniles, 84.8 cc. in 15 adult females, and 102.1 cc. in 10 adult males. 
As best shown by figure 2, the cranial capacity increases very rapidly 
at first, but continues to gain at a very much slower rate during the 
later part of juvenile life. In its relation to body weight (in g.) the 
cranial capacity (in cc.) decreases during growth enormously until 
about the middle of the juvenile period. Thereafter this percentage 
relation diminishes more gradually with increasing body weight, i. e., 
the larger a specimen the proportionately smaller is its braincase. 
This relative cranial capacity averages in Nasalis 7.01 in infants, 2.23 
in juveniles, 0.88 in adult females, and only 0.51 in adult males. The 
marked sex difference in adults is mostly, if not entirely, due to the 
striking sex difference in general body size. From figure 2 it is evident 
that the relative size of the cranial capacity of Nasalis does not differ 
significantly from that of langurs and macaques of corresponding body 
weight.^ These lower catarrhines stand far below the great apes in 
regard to their relative cranial capacities (see also Schultz, 1941, fig. 4). 

The relative face height (total face height in percentage of trunk 
height) decreases in general during growth, i. e., the height of the face 
increases at a slower rate than the height of the trunk. This is a general 
rule among primates, as far as is known so far. In male proboscis 
monkeys the face becomes proportionately higher again late in growth, 
undoubtedly in connection with the final phases of development in 
their large dental apparatus. 

The first two indices in table 4 dealt with the relations in size be- 
tween the brainpart as well as the facepart of the head and the trunk. 
It was shown that the first of these relations changes with age very 
much more than the second. From this it can be anticipated that the 
proportion between the size of the face and the size of the brainpart 
of the head also changes during growth. The latter relation is ex- 
pressed precisely by the relative upper face height (distance between 
nasion and stomion in percentage of average head diameter). This 

'As it happens, all the records for 5 langurs (different species) fall slightly below the curve 
for Nasalis and the data for 5 out of the 6 macaques lie a little above this curve. It requires many 
additional data before it can be decided whether there exists here a real, though at best slight, 
difference. 



SCHULTZ: THE PROBOSCIS MONKEY 



291 



I 

> 






0) 



en 

C 

o 



o O 



o 


J2 


a 


^ 




rn 


(1) 




^H 


cr; 


+J 


fl 






c 


o 


• ^* 


o 


en 


Tl 


0^ 




M 


bC 


C 


C 








S 


be 


o 


< 


' ^ 




r/1 




C 




O 



Relative 
Ear Size 


^2 


t- ^ 


x2 


■^ ^ 


»0 05 


'^ J: 




^•^ 


«5t^ 


^;i 


%~ 












e 














05 (M 


c 


<^^ ?< 




2^ 




S 00 


2t^ 


2i^ 


2tl 


5^ <33 


^ >-< 


■•^ .-H 


I—! 


i—i 


T— t 


1—1 


s ■ 












►~H 














• 










g t^ 


t> 


_ 


, 


_ l^ 


^ CO 


—1 ^ 


^ u- 





05 CO 


TfH t^ 


e ^ 


CO 1 


1 


-*; I 


<=> J 


1 


^►^ 




^^ 




^S; 


=^ 


u 














r~^ 


^ "^ 


lo 





1—i 


fO CI 


^ OC 


fo OC 


--^ Oi 


<N 


"* ^ 


cd 1 


' 


CO 1 


-* ►! 


"^ X 


1) v^ 


^S2 


"^ t> 


^ X 


cot: 


00 § 


o 












■» 












;j 
















7^ 





CO ?2 


p 00 


13 S- -52 


^'k 


^ A 
-* 5 




§§ 




KJ 












^ •*-A 


t^ 


t^ 


_ 


►. 00 


^ <=> 


■^ S "& 


t> (M 


-- o^ 


^ C^ 


t^ r-1 


OC c^i 








-^ 


• 1 

1— ( 


2o^ 

r-1 


« 














.— f 


•* 





^ 05 


00 


•5 cyj 


(N lO 


00 -^ 


t^ (M 


-* T-H 


^ r-H 


Rela 
lead 




Ss 


^ 2 


i-H 


^2 


1 












CO 












g^ 












^ 












fe 


<r> 


CO 


K 


»C 





'S 




1— ( 




I—" 


1—1 


4^ 












?i. 












CO 























^ 


|» 


to 


cc 










^ 
s 


C 
e3 




0; 

3 


;§ 
3 






e.- 


^ 
2 


-a 


T3 




fo 


hH 


>-? 


< 


< 



292 bulletin: museum of comparative zoology 

index increases in proboscis monkeys from an average of 40 in fetuses 
to one of 80 in adult males. The rise in this index is greatest after 
infantile life when the braincase has attained most of its final size, 
while the face grows most intensively under the influence of the greatly 
expanding second dentition. 

The cephahc index (greatest head breadth between temporal 
muscles in percentage of greatest head length from glabella) of the 
proboscis monkey decreases during the early growth period, only to 
increase again during juvenile life. The latter rise in this index is 
caused by the strong development of the temporal muscles accom- 
panying the completion of the permanent dentition. The shape of the 
braincase itself becomes steadily narrower in relation to its length, as 
is shown by the following averages for the cranial index (greatest 
breadth of braincase above supramastoid ridges in percentage of 
greatest mid-sagittal length of braincase from glabella) : infants = 80.3, 
juveniles =77.5, adult females=72.4, adult males = 69.7. A corre- 
sponding decrease in this index has been found in chimpanzee and 
in orang-utan (Schultz, 1940a, 1941). The relation between the 
height and the length of the skull (greatest skull height from basion, 
perpendicular to ear-eye horizon, in percentage of skull length from 
glabella) also decreases with age in the proboscis monkey, i. e., the 
skull becomes gradually lower in relation to its length. This is demon- 
strated by the following averages: infants = 64.8, juveniles = 64.5, adult 
females = 60.5, adult males = 57.9 (see also fig. 7). In the orang-utan 
this index remains constant during postnatal growth (Schultz, 1941). 

The face index (upper face height in percentage of face breadth 
between zygomatic arches) increases very markedly with advance in 
age not only in the proboscis monkey (table 4), but also in orang-utan 
and chimpanzee (Schultz, 1940a, 1941) as well as in man. The in- 
terocular index (breadth between inner angles of the eye-clefts in per- 
centage of face breadth) of proboscis monkeys decreases with growth 
from fetal to infantile life and increases again thereafter, particularly 
in males. These ontogenetic changes correspond exactly to those 
occurring in orang-utan and chimpanzee (Schultz, 1940a, 1941). 

The last index in table 4 shows the relative size of the outer ear 
(ear height X ear breadth in percentage of head length X total head 
height) in proboscis monkeys of different ages. According to this index 
the outer ears become proportionately larger until infantile hfe (see 
also figure 3), to decrease again in relative size during later stages of 
growth. This is precisely what happens also in man, orang-utan and 
chimpanzee (Schultz, 1933a, 1940a, 1941). 



SCHULTZ: THE PROBOSCIS MONKEY 



293 



ding averages of 
ithor (1937), but 
n Borneo on the 


Asiatic Primate Expedition of 1937. 


9Z%S 

.m^ py 


Tf IC 

00 t>; 


12.4 
11.9 


14.1 
13.3 


t^ p 


14.2 
16.2 


xapuj 
j.irinooj,9juj 


19.7 
21.8 


21.5 
21.3 


22.4 

21.7 


14.7 
13.7 


15.7 
16.7 


compared with correspon 
a former paper by the au 
collected and measured i 


uaddn -jay 


69.3 
80.0 


56.2 
.53.7 


56.7 

.58.1 


78.7 
87.3 


71.3 
73.6 


j¥ph 

30OJ -py 


i> 00 

d 00 

T—>. 1—1 


1 — 1 i-H 


^ oo 
1—i .—1 


21.4 
22.6 


19.2 
21.4 


9Ztg 


T— 1 T— 1 


C5 r-; 
t^ 00 


CM 00 


21.8 
21.8 


20.4 
21.8 


onkeys 
re from 
ecimens 


xapuj 


121.0 
121.2 


104.4 
102.5 


109.3 
110.5 


p !>. 

d CC' 

1-^ T-H 


r-H 05 

^ CM 

<— ( I— 1 
1—1 1— I 


able 5 

boscis m 
lulatta a 
from sp 


mSuaj qimq 
Mddn •p'Q 


132.2 
137.6 


120.5 
123.5 


CO t^ 

00 CM 


P P 

.-H d 


1—1 T—t 

1—1 y—i 


T 

ions of adult pro 
ata for Macaca d 
1 other data are 


il0udq qmiq 
dacnoj py 


109.1 
113.6 


1 15.3 
120.2 


99.1 
101.9 


96.1 
97.6 


99.5 
99.0 


mdudq 


152.3 
156.6 


200.2 
197.3 


195.5 
201.2 


186.8 
181.1 


68.2 
72.8 


proport 

The d. 

ales. Al 




108.0 
125.0 


102.9 
111.0 


102.4 
108.8 


103.5 
108.8 


97.4 
110.4 


some body 

macaques. 

two new m 


su9mp9dg 


lO o 


m lO 


t^ CO 


LO o 


1— t 


X9S 


o ^ 


o 'b 


O IV) 


O 1^ 


o 'b 


Averages of 
adult langurs anc 
include those for 


60 


> 

.23 


d 

g 
X 

1 




»3 

3 
,y» 

c3 


33 

-^ 

r- 

c 

o3 
si 

Zj 

o3 



294 bitlletin: museum of comparative zoology 

Table 5 gives a comparison between some of the body proportions 
of adult proboscis monkeys and corresponding averages for adult 
langurs and macaques. From these data it is seen that the proboscis 
monkey has a relatively stouter trunk than either the langurs or the 
macaques. The tail of the proboscis monkey is proportionately much 
shorter than that of langurs and of the crab-eating macaque. The 
proboscis monkey is furthermore characterized by the relatively 
longest upper extremities, highest intermembral index and propor- 
tionately smallest outer ears. With the langurs the proboscis monkey 
has in common a comparatively small brainpart of the head and a 
broad interocular region, in contrast to macaques in which the ap- 
proximation of the eyes has gone to a greater extreme. If the figures 
in table 5 are compared with the corresponding data for adult man- 
like apes and man, as published by the author in former papers (1933a 
and c, 1940a, 1941), it becomes very evident that the lower catarrhines 
are really remarkably uniform in most of their body proportions and are 
far removed in these respects from all the higher primates. 



DEVELOPMENTAL CHANGES 
Outer Body 

The smallest fetus of A^asalis that has been briefly described is the 
specimen of Keibel (1906) with a crown-rump length of 25.2 mm. 
(see fig. 3). At this early stage of development the nose is already 
slightly more prominent than in other primate fetuses of corresponding 
age. The eyelids have not yet closed, the first signs of hair anlagcn have 
appeared on the eyebrows, upper lip and chin. The tip of the tail bears 
a caudal filament, as it does also in the older specimen of Keibel and 
in the four youngest fetuses, studied by the writer. The youngest of 
the latter are the double-ovum twins, collected on the Asiatic Primate 
Expedition of 1937, with sitting-heights of 119 and 124 mm. respect- 
ively. In both these specimens skin color has appeared on the scalp 
(grey-brown), the already quite prominent nose (deep slate-blue), the 
digits (light brown) and, in the male, the perineal raphe (grey to blue), 
the rest of the body surface being still unpigmented. Of hair these 
two fetuses possess dark, long eyebrows and light as well as dark hairs 
on the upper lip and chin (fig. 3). Ischial callosities are already clearly 
indicated by sharply circumscribed, thickened zones overlying the 
ischial tuberosities. The outer ears are still of moderate size, the edge 
of the helix has not yet become rolled in, they bear a much more 



SCHULTZ: THE PROBOSCIS MONKEY 295 

marked Darwinian tubercle than persists later on, and they possess a 
definite lobule. The same conditions exist in the next older fetus with 
a sitting-height of 133 mm., except that very fine and short, dark hair 
has been added on the scalp, where it radiates from a center on the 
forehead and is directed straight back over the parietal and occipital 
regions. Even at this stage there are as yet no eyelashes. In another 
fetus, measuring 154 mm. in sitting height (fig. 3), the hair on the scalp 
has become very much longer, is parted on the forehead, and has 
assumed a reddish-brown color. The outer sides of the limbs and the 
back have acquired short, reddish hairs, of which there are fewer on 
the lower than on the upper extremities. In the older fetuses the hairy 
coat has been completed. It may also be mentioned that in all the 
fetuses examined by the author, as well as in the infants, the nose is 
conspicuous not only on account of its size and shape (fig. 3), but also 
by its deep blue color. Wiedersheim (1901a and b) had made the 
same observation on his fetus of Nasalis, but expressed the opinion 
that this blue color might be some post mortem change connected with 
preservation. This, however, is not the case since this color was clearly 
noticeable in the fresh fetuses, removed by the writer very soon after 
the death of the mother animal, and also in several living infants, 
observed in the field. Later during postnatal growth this blue color of 
the nose fades away to change to a more or less deep shade of brick- 
red, probably due to the high state of vascularization of this specialized 
structure. In adult proboscis monkeys there exists a very striking sex 
difference in the relative size of the outer nose, as is clearly shown by 
plates 1 to 3. The nose of the adult female has not developed far 
beyond the stage reached by all infants, but in adult males this organ 
is several times larger than in the young or in the fully grown female 
and the lower profile of the nose is horizontally directed, instead of 
forward as in the females. In old males, finally, the enormous nose may 
become somewhat longer yet and sag to an extent that it overhangs the 
mouth. Incidentally, the writer has never seen a proboscis monkey 
with as markedly bifid a nasal apex as claimed as typical and illustrated 
(diagrammatically) by Pocock (1926). 

Among all the proboscis monkeys studied there occurred only one 
abnormal condition, noticeable on the outer body^ This is the case of 
an incomplete supernumerary digit which had developed as a grape- 
like appendage from the base of the terminal phalanx of the right 

^Reference may here be made also to the case of unilateral cryptorchism, found in one 
adult of this series, in which one testis had been stunted in development and had failed to 
descend (Schultz, 1938 b). 



296 bulletin: museum of comparative zoology 

second toe in an adult male specimen (Plate 4, D). A practically 
identical condition was found in an adult male langur from northern 
Siam (A. P. E. No. 121) in which there also existed a grape-like 
appendage, bearing dermatoglyphics, but in this case it had sprouted 
from the middle segment of the left second finger. This is the only 
incidence of this sort among 83 langurs of various species, examined 
by the writer. In macaques developmental abnormalities of the digits 
appear to be somewhat more common. In a total of 152 specimens, for 
which the writer has kept records, incomplete supernumerary digits 
of the same appearance as the just mentioned cases were found in a 
Macaca irus (ad. cf , A. P. E. No. 427) on the basal phalanx of toe II 
and in a Macaca mulatta (ad. cf , A. S. 1541) on the terminal phalanx 
of toe III. In another Macaca irus (ad. cf", A. P. E. No. 369) there are 
small supernumerary digits on the basal phalanges of the right toes III 
and IV and in the left fourth toe the basal and middle phalanges had 
very incompletely developed, leaving this digit abnormally short. 
The latter condition is nearly duplicated in a juvenile Macaca mulatta 
of the Carnegie Colony, though in this case it is the fifth toe. In 
gibbons such manifestations of abnormal development are surprisingly 
common, as will be shown in a later publication. 

Teeth 

The sequence of eruption of the deciduous teeth of proboscis mon- 
keys could be studied on the basis of the findings in the incomplete 
dentitions of five infants. The following sequence appears to represent 
the typical condition: upper middle incisors, lower lateral incisors, 
lower middle and upper lateral incisors, lower and upper first molars, 
lower canines, upper canines, lower second molars, upper second 
molars. This order of appearance differs only in some details from the 
sequence of eruption of the deciduous teeth of macaques in which all 
the incisors are also the first to appear, being followed after a con- 
siderable pause by the canines and first molars and, after a long resting 
period, by the second molars (Schultz, 1933 b). In orang-utan and 
chimpanzee, on the other hand, the canines are the last milk teeth to 
erupt (Schultz, 1940 a, 1941). Modern man stands in this respect 
much closer to the lower catarrhines than to the great apes. 

The sequence of eruption of the permanent teeth of proboscis 
monkeys was studied by means of the dental formulae of 15 specimens, 
listed in table 6, considering also the degrees of eruption of particular 
teeth. The order of appearance of these teeth is as follows: lower first 



SCHULTZ: THE PROBOSCIS MONKEY 



297 



Table 6 

Dental formulae of 15 proboscis monkeys with incomplete perman- 
ent dentition, arranged to show the sequence of eruption of the per- 
manent teeth. Seven of these cases are from former paper by the 
author (1935). u. = tooth of upper jaw, 1. = tooth of lower jaw. 



Specimens 



Dental Formula 



Incompletely Erupted: 



1 9 



1 c? 



1 ? 1 o^ 



2 9 2 a" 



i i c m m M 



i i c m m M 



i i c m m M 



I i c m m M 



I - c m m M 



I I c m m M 



I I c m m M 



I I c m m M 



I I c m m M M 



I I c m m M M 



I I c m m M M 



II C m m M M 



I I c P P IM M 



I I C P P M M 



I I C P P M M 



I I C P P M M M 



1 1 C P P M M M 



1 1 C P P M M M 



In 1 case u. Ml (M)-and 1. Ml (H) 



1. II (M) 



1. 12 (H 



u. 12 (M) 



u. and 1. 12 (3^) and u. M2 (M) 



l.C(^) 



l.Cf^ 



9 =u. C (H) and 1. M3 (J4) 
c^=u. C (K)andl. M3 (^) 

1 9 =1. M3 (H) and u. MS (M) 
1 9 =u. M3 (H) 

1 cf =1. & u. M3 (H) and 1. C (%) 
and u. C (K); 1 o^ =1. C (%) and u. 



298 bulletin: museum of comparative zoology 

molars, upper first molars, resting period, lower middle incisors, upper 
middle incisors, lower lateral incisors, upper lateral incisors, lower 
second molars, upper second molars, lower and upper first and second 
premolars in unknown, but probably very rapid and irregular succes- 
sion, and finally, lower canines, upper canines, lower third molars, 
upper third molars in females, and lower third molars, upper third 
molars and lower canines (simultaneously), upper canines in males. 
This sequence agrees more closely with the corresponding conditions 
in macaques and baboons than with those in langurs, inasmuch as the 
second molars erupt after the incisors and the third molars after the 
premolars, instead of before, as in langurs (Schultz, 1935). 

The type of occlusion of the incisors was noted in the freshly killed 
specimens and later again in their macerated skulls and this with 
identical results. In infants edge-to-edge bite occurred in 4 cases and 
underbite in 5 cases, in juveniles the former condition was found in 
4 cases and the latter in 3 cases. Among adults edge-to-edge bite 
existed in 6 cases and underbite in 20 cases without any significant sex 
difference in this regard. Overbite did not occur at all in any of the 
proboscis monkeys examined, but it is very common among macaques 
and not rare in orang-utans and chimpanzees. 

For the sake of completeness the abnormal and pathological con- 
ditions in the dentition of Nasalis are briefly recorded here. For this 
purpose the data on 48 specimens, used in a former study by the author 
(1935), have been combined with the records on the 42 new skulls, 
examined for this paper. There are no pathological or abnormal con- 
ditions in the dentitions of 9 infants and 14 juveniles. Among the 67 
adults the lower left middle incisors are congenitally lacking (leaving 
no diastema) in 2 cases and crowding of teeth exists in 2 other cases, 
affecting the upper middle incisors, once bilaterally and once uni- 
laterally. In the same series of adults no case of carious infection 
could be observed, but alveolar abscesses exist in 2 skulls (once on a 
premolar and once on a canine) and closed alveoli, following the 
loss of a tooth, are present in 2 other skulls (once upper r. and 1. lateral 
incisor and once only the right one of these teeth). In comparing these 
notes with the author's (1940a, 1941) corresponding reports on orang- 
utans and chimpanzees, it will be strikingly apparent that abnormali- 
ties and diseased conditions of the dental apparatus are far more 
prevalent in these anthropoid apes than in the proboscis monkeys. 



SCHULTZ: THE PROBOSCIS MONKEY 299 



Skeleton 

a. Spinal Column. The numbers of vertebrae in the different 
regions of the spinal column vary comparatively little in proboscis 
monkeys and certainly not nearly as much as in anthropoid apes. In 
the 46 specimens of Nasalis, for which the author has been able to 
collect the vertebral formulae, all have 7 cer\acal vertebrae. In 41 of 
these specimens the presence of absence of transverse foramina in the 
cervical vertebrae could be noted. These foramina occur in all cervi- 
cal vertebrae, except in the seventh, where they exist bilaterally in 
29 cases, unilaterally (on right side only) in 4 cases, and are completely 
lacking in the remaining 8 cases. In macaques these foramina are 
lacking in nearly all cases, in orang-utans in all cases, in chimpanzees 
in almost half the cases, but in man extremely rarely. 

In the 46 Xasalis skeletons there are, on an average, 12.1 thoracic 
vertebrae, 13 occurring in 4 cases and 12 in all the other cases. In the 
specimens with 13 pairs of ribs the last pair is always very short, in- 
deed, in two instances the thirteenth ribs are mere rudiments. The 
average number of lumbar vertebrae equals 6.9, 7 being present in 41 
cases and only 6 in 5 cases. Six lumbar vertebrae combine with 13 
thoracic ones in 4 instances and with 4 sacral vertebrae in the fifth 
case. Only in the latter case are there 18 thoraco-lumbar vertebrae^ 
and in all the other specimens of this series there are 19 thoraco- 
lumbar vertebrae. This remarkable scarcity of numerical variations 
among the praesacral vertebrae of proboscis monkeys forms a striking 
contrast to the conditions in anthropoid apes in which these vertebrae 
vary to an enormous extent (Schultz, 1930). The average number of 
sacral vertebrae of the proboscis monkey is 3.1, the sacrum being 
composed of 3 vertebrae in 42 cases and of 4 vertebrae in the remaining 
4 cases. One of the latter cases is shown in figure 4. This figure serves 
to illustrate also the extreme variations in the position of the sacrum 
in relation to the hip bones. In general, the sacrum is attached higher 
on the ilia in specimens with only 6 lumbar vertebrae than in those 
with 7, though even among the latter there exists considerable varia- 
bility in this respect. An even greater variability in this relative posi- 
tion has been found in orang-utans in which it is also correlated to a 
considerable extent with the number of praesacral vertebrae (Schultz, 
1941). The number of caudal vertebrae of proboscis monkeys averages 

iPlower (1888) gives the vertebral formula of a Nasalis skeleton in the Royal College of 
Surgeons in London as 12 thoracic, 6 lumbar, 3 sacral, and 27 caudal vertebrae. 



300 



bulletin: museum of comparative zoology 



24.6 and varies between 23 (4 cases) and 26 (5 cases). The total num- 
ber of vertebrae varies between 52 and 55, averaging 53.7. The typical 
vertebral formula of Nasalis reads: 7 cervical, 12 thoracic, 7 lumbar, 
3 sacral, and 25 caudal vertebrae. The average vertebral formulae of 
nearly all other lower catarrhines are exactly the same, except of course 
in regard to the caudal region. This formula, disregarding the caudal 
region, occurred in 85 per cent of the proboscis monkeys. In the orang- 




A. PE. 293 



A.RE. 287 



Fig. 4. Pelves of adult male proboscis monkeys: specimen on left has 7 
lumbar and 3 .sacral vertebrae, specimen on right 6 lumbar and 4 sacral 
vertebrae. Note difference in relative position of sacrum. 

utan the typical formula for the praecaudal spine was found in only 24 
per cent of the (105) specimens and each of the many different formulae 
is even less frequent (Schultz, 1941). The corresponding figure for 
chimpanzee' is 32. 

The proportionate lengths of some of the various spinal regions 
undergo significant changes during growth in most primates, as has 
been shown in other studies by the author (1938 a, 1940 a, 1941), for 
which, however, no very young specimens of Nasalis had been avail- 



'Based upon data on 77 chimpanzee skeletons, published by the author (1940 a). 



SCHULTZ: THE PROBOSCIS MONKEY 



301 



able. Data for the latter have since been secured and are listed in 
table 7 together with figures from a previous paper by the writer (1938 
a). The relative lengths of the cervical and thoracic regions become 
shorter with age and the lumbar region increases in its proportionate 
length. This is in complete agreement with the findings for the other 
primates. The relative length of the caudal portion of the spine reaches 
its maximum values during the infantile period of growth, as was to be 
expected from the identical age changes in the relative tail length, as 
measured on the outer body (table 1). 



Table 7 

Averages of the lengths of the various spinal regions (measured with 
tape on the ventral surface) in percentage of the trunk height in 
proboscis monkeys of different ages. From author's former paper 
(1938) and new data for fetus and for two small infants. 





Speci- 








Thor.- 






Age 


mens 


Cerv. 


Thor. 


Lumh. 


Lumb. 


Sacr. 


Caud. 


Fetus 


1 


23.2 


49.3 


40.6 


89.9 


13.8 


149.3 


Infants 


9 


20.6 


47.8 


42.0 


89.8 


15.1 


189.5 


Juveniles 


6 


17.4 


42.2 


45.0 


87.2 


14.5 


177.9 


Adults 


23 


17.5 


42.5 


44.9 


87.4 


13.4 


160.1 



The available material permits the following observations and con- 
clusions regarding the ossification of the vertebrae in proboscis mon- 
keys: The two ossified lateral portions of the dorsal arches become 
united in the mid-line at birth or even sooner, though the dorsal proc- 
esses remain cartilaginous for some time later. The only exception to 
this was found in the second largest infant, with complete and some- 
what worn deciduous dentition, in which the dorsal arch of the atlas 
remained open (Plate 4, C). This is undoubtedly an anomaly and may 
be regarded as localized spina bifida occulta. The vertebral bodies and 
the dorsal arches are still separate in only the three youngest infants 
with incomplete first dentition. The bony union between these verte- 



302 bulletin: museum of comparative zoology 

bral elements is complete at the time the last deciduous teeth have 
appeared, i. e., much earlier than in the orang-utan (Schultz, 1941). 
The odontoid process of the epistropheus becomes fused with the body 
of this vertebra at the age of eruption of the .first permanent molars 
and the ventral arch of the atlas unites with the lateral portions of the 
first vertebra very soon afterwards, or also at a comparatively earlier 
age than in the orang-utan. The sacral vertebrae fuse late in the 
juvenile period, the first and second somewhat sooner than the second 
and third and the dorsal side before the ventral side. The epiphyseal 
plates of the vertebrae have completely fused with the vertebral bodies 
in only two old proboscis monkeys. In all the other specimens they are 
still separate at least in part of the thoracic region. These epiphyses 
close first on the caudal vertebrae, namely at the age of completion of 
the permanent dentition. During later adult life more and more of 
these plates fuse with the corresponding vertebral bodies and this 
apparently in an order starting simultaneously at the opposite ends of 
the praecaudal spine. 

b. Sternum and Ribs. The length of the sternum (without xiphoid 
process) in percentage of the trunk height averages in proboscis mon- 
keys 22.9 in 9 infants, 17.9 in 7 juveniles, 17.6 in 15 adult females, and 
22.2 in 10 adult males. The sex difference in adults is significant as 
the index varies in the females between 15.0 and 21.3, whereas in the 
males between 20.0 and 25.3. The corresponding averages for adult 
orang-utans, chimpanzees and, particularly, man are considerably 
higher, indicating a proportionately longer sternum than in proboscis 
monkeys. 

The number of pairs of ribs reaching the sternum could be recorded 
in 41 proboscis monkeys and this invariably as 7. In many other lower 
catarrhines 8 pairs are the rule. The slender sternum of the proboscis 
monkey ossifies from a number of centers, as shown in figure 5. Double 
centers in an intercostal space, frequent among higher primates, have 
never been seen in proboscis monkeys. The corpus sterni contains four 
centers in all infants and juveniles, but a fifth center may appear later, 
as it was found only among adults, namely in 12 out of 25 cases. Only 
in two old animals have some of these segments become partly fused, 
once the lowest two and once the lowest three segments. The xiphoid 
process is ossified in many specimens, containing a center already at 
birth (fig. 5), but remains permanently cartilaginous in a minority of 
the cases. 

A separate epiphysis at the tubercle of the ribs disappears during 
juvenile life, but the epiphysis of the costal head remains at least 



SCHULTZ: THE PROBOSCIS MONKEY 



303 



partly separate for a considerable time after the completion of the 
permanent dentition. 

c. Hip Bone. According to X-ray photographs of the proboscis 
monkey fetuses the innominate bone has only its three primary elements 
ossified during prenatal development and these are still widely sepa- 
rate. Even the bony rami of ischium and pubis do not become closed 
until about the time of birth. At the junction of the pelvic elements in 
the hip joint there is a separate os acetahuli in 4 out of 7 juveniles. 
These elements fuse during the last brief phase of dental eruption, the 
pubis remaining separate slightly longer than the ilium and ischium, 




NEWBORN 





ADULT cf 



Fig. 5. The sternum in proboscis monkeys of different ages. Dotted areas 
cartilage. 



the acetabular lines of separation for the pubis being still visible in 
one fully adult specimen. This agrees very closely with the correspond- 
ing findings for the great apes (Schultz, 1941), but differs from these 
conditions in man in whom the pehac elements fuse long before the 
second dentition is completed. The epiphyses on the ischial tuberosi- 
ties loose their separate existence early in adult life, but the bony rims 
on the iliac crests remain separate, at least in their middle portions, 
for a considerable time after adulthood has been attained. The pubic 
symphysis was found open in all specimens, except one old male in 
which it had become solidly closed along its uppermost third. 

d. Shoulder Girdle. The sternal epiphysis of the cla\dcle does not 
become fused until adult life. It is uncertain just when it does close in 



304 bulletin: museum of comparative zoology 

relation to other happenings in skeletal development, as it was closed 
in one adult before many epiphyseal lines on limb bones had disap- 
peared, but was still open in several other adults with all the epiphyses 
of the limb bones closed. 

The coracoid process of the shoulder-blade develops an ossification 
center shortly before birth. This process remains separate until the 
second dentition is completed. A separate bony epiphysis on the 
acromial process does not always disappear until somewhat later, i. e., 
in adults with noticeable attrition of the teeth. The vertebral border 
of the scapula is mainly cartilaginous in all juveniles and the lower 
angle remains cartilaginous throughout life. Finally, it may be men- 
tioned here that one adult male proboscis monkey possesses very pro- 
nounced scaphoid scapulae. 

e. Limb Bones. The first ossification center to appear in the epi- 
physes of the long bones of the limbs is the one for the distal epiphysis 
of the femur which is already of considerable size at birth (fig. 6). It 
is very remarkable that no other epiphyses of the limb bones have de- 
veloped ossification centers in the newborn proboscis monkey, since 
most of these centers are already present in newborn macaques 
(Schultz, 1937). In a proboscis monkey infant with merely the second 
milk molars lacking, ossification centers have appeared in the epiphy- 
ses of the long bones, except the medial epicondyle of the humerus, 
the proximal epiphysis of the radius, both epiphyses of the ulna, the 
trochanters, and the proximal epiphysis of the fibula. Of all the latter 
epiphyses, that at the proximal end of the radius develops first an 
ossification center, namely as soon as the deciduous dentition is com- 
plete. Slightly later the center for the great trochanter is added, then 
those for the medial epicondyle and the proximal epiphysis of the 
fibula, to be followed by a center for the small trochanter and last by 
one for the proximal epiphysis of the ulna which was still absent even 
in two specimens from the beginning of the juvenile period. The 
epiphyses of the metatarsi, metacarpi and phalanges develop centers 
in early infantile life (fig. 6), but the epiphysis of the tuber calcanei 
does not become ossified until the first permanent molars have erupted. 
In the patellae ossification centers appear toward the end of the in- 
fantile period. 

The first of the epiphyses to become solidly united with the shaft 
of the bone is the distal epiphysis of the humerus which disappears as 
a separate element just before the permanent dentition is completed. 
After an appreciable interval the medial epicondyle of the humerus and 
the proximal epiphysis of the ulna fuse with the shafts of the bones. 



SCHULTZ: THE PROBOSCIS MONKEY 



305 



a -= 



V 


c 




-o 






;-i 






C} 


«4-l 






O 




-kJ 


rr 




a 


s 




a 


s 


c 




a> 


.j; 


c3 


^ 


OJ 


G 


o3 






S 


-3 




!/2 


a 

=3 


-c 


O 


§ 




u 


t- 


h 












c 


:rt 




""" 


p 


0) 


I-, 


fl 


43 


o 



- t^ o 

i, C i£ 

5 § « 

3 S ^ 



^' o 



tE 



33 

;h 

O 

- ^ = 

o £. o 

a - c 
2 S g 

c3 



5= -§ «^ 






- 5; 5 



O C3 .= 

S t, ■ - 

« « <a 

. -*^ c 



25 






-1-3 

3 3 

J - 
5^ 3 



« c 

— c 

S3 ?? 

— 03 







(1) 




















































>5 




















X 






T3 

























>i 


C 




, 




T3 


«2 




>< 

o3 






;h 


'7. 

s 
J 







-5 


■J. 

_c3 

1—1 


5 

"0 



3 

1-— 


2 


>< 


a: 


3 

s 


3 

3 
l-M 






a: c 






?; =^-^ ? 



-a 
'" 3 



33 

o 



— c 



- 9- 



p; -^ :^ i^ 






.^^ -S "3 S 

-2 _C ^ X' — 



H pM 



-a 

3 



o3 



o 
i-, 






o "> 

c3 c: 



O 



3 



_0 



O 



- II.- 



~ t> Jl. g 



— < o3 

2 ;:-- 



03 



■Ji 

3 



3 3 3- 

3 .t. _tA '^ 



3 



32 

■3 



CO 

02 
<U 

3 g 

3 '" 
32 






o 



03 r^ O 

^ "^ ^ .5 -3 



c3 
c3 ^3 



03 o3 

a; o 
S 2 < O H E^ 



3 ^ 



5< o a £ 



p 2 

X 3 

3 ■*^ 

11 



-3 3 



03 



O > 

c3 o3 



I— 1 1-^ I— I )— I I— I ^ 









y 


















^ 


. 




>. 






^ 


■s. 




•^ 






i:: 




^ 




•-^ 




•J. 


>< 


p 


a: 

3 


1 


!- 


'^m 


-^ 


c 











r- 


^ 




7i 


Z^ 



3 ^ 



P=5 

















CO 
02 














^ 


-1^ 


p 


X 


a. 


X 


P 


-i 



p 


CO 


'7. 


^.^ 


a; 
3 


a: 


03 


02 

Si 

5 






— .J 
1—1 


c3 


r= 


r— ^ 


o3 
IS 


Acetab 
Coraco 


3 
S 


03 

c3 
-^ 
V 

<5 


c3 
S 

2 


S 

a 
3 

h-i 


a3 
3 


-73 
03 


3 



92 






5 X 



2 "3 



- c3 . "~ := 

c3 -3 -3 03 C 

~ ~ O 3 -;3 

-- x »3 1-^ -" 

v^ I— I >^ »-^ ^ 



■3 •:■ 

03 ? 



" — i " r" 

t %■■% -3 3 

S g ^ a £ 

^ iL. p:; H t^ 


























. 


X ^ 




-*i 


r^ 


,-^ 






3 


X 


CO 


a; 


CO 


a: 
3 


<a 


a2 

2 


-*- 







o3 3 


w 


3 
03 








CJ 


a; 


J 
« 

2 


IS 


1 s 


c3 


g 
3 


> 



_c3 



306 



bulletin: museum of comparative zoology 



The further, detailed sequence of epiphyseal union in the proboscis 
monkey is shown best in table 8 where some other, previously dis- 
cussed items of skeletal development have been inserted. From the 
preliminary comparison of these conditions in Nasalis with those in 
a few other primates, as far as known, it becomes apparent that, in 
general, there seems to prevail one ground plan for the sequence of 
these osteological happenings. More than detailed exceptions are 
found chiefly in regard to the order of fusion in the coracoid process, 




'^ffl 



A.FE. 232 



A.RE. 235 



Fig. 6. Tracings of X-ray photographs of the limbs of infant proboscis 
monkeys. A = practically "newborn" (not over two weeks old), B = com- 
pleted deciduous dentition (Ml nearly ready to erupt). 



ischial tuberosity and the primary pelvic elements, which stand near 
the top of the list for Nasalis, whereas far down in the lists for the 
great apes and high and low in the list for man. It must be mentioned 
here that most of the items listed in table 8 fuse after the completion of 
the permanent dentition in the monkey and the apes, whereas in man 
nearly all the items have already fused at that stage of dental develop- 
ment. Further comments on these significant problems are best post- 
poned until the corresponding data on many other primates have been 
fully recorded. 
/. Skull. In a newborn proboscis monkey the great or bregmatic 



SCHULTZ: THE PROBOSCIS MONKEY 307 

fontanelle is still wide open, extending far forward between the frontal 
bones. This is quite surprising inasmuch as these fontanelles are 
either diminutive in size or else already closed at birth in macaques, 
chimpanzees and orang-utans (Schultz, 1937, 1940 a, 1941). The lamb- 
doid and pteric fontanelles are closed in all infant proboscis monkeys, 
but the asteric or mastoid fontanelles are of considerable size at birth 
and their lower portions, near the lateral occipitals, remain open 
throughout the infantile period, i. e., long after the bregmatic fon- 
tanelle has disappeared. Indeed, in one adult Nasalis both mastoid 
fontanelles have failed to close as a developmental anomaly (Plate 4, 
B). The pteric fontanelles are closed in fetal life in such a fashion that 
the alisphenoid meets the parietal, thus keeping the temporal squama 
from meeting the frontal. This arrangement existed in all 38 Nasalis 
in which the sutures of the pteric region had not yet become obliterated 
(see fig. 7). 




NEWBORN 



Fig. 7. Side views of skulls of proboscis monkeys, posed in ear-eye horizon, 
reduced to same skull length (from glabella) . 

Of other notes on early cranial development in Nasalis the follomng 
are here recorded: A cranio-pharyngeal canal, present in all embryos, 
did not persist to postnatal life in a single case, though it is found in 
many other primate skulls. The mandibular symphysis is still open 
in a newborn and one infant, lacking its second molars, but it is closed 
in all other infants including two with incomplete first dentition. The 
temporo-mastoid sutures disappear very soon after birth. The inter- 
frontal or metopic suture is entirely open in the two youngest infants 
and is partly open in 6 out of 7 other infants, closure beginning in- 
variably in the middle portion of this suture. In exceptional cases the 



308 bulletin: museum of comparative zoology 

metopic suture can persist beyond the infantile period. This has hap- 
pened in one juvenile (A. P. E. No. 248) with complete metopism and 
in an adult female (U. S. Nat. Mus. No. 196792) among the N^asalis 
skulls examined by the author. Another abnormal suture was found 
in one adult proboscis monkey, namely a perpendicular suture dividing 
the right and the left zygomatic bone (Plate 4, A). The occipital ele- 
ments are separate in all infants and fused in all juveniles, i. e., the 
basioccipital, the lateral occipitals and the occipital squama become 
solidly joined at the time of eruption of the first permanent molars. 
This agrees perfectly with this condition in the great apes and in man 
(Schultz, 1941). All the other sutures remain open until at least the 
beginning of adulthood. The sequence of suture closure is difficult to 
determine since there exists little detailed consistency, as shown by 
the data in table 9. All that can be concluded from the available ma- 
terial is that the lambdoid and sagittal sutures generally close before 
the coronal and basilary sutures, and that the coronal suture is closed 
only in old animals with an advanced degree of attrition of the teeth. 

Of the remaining sutures it may here be mentioned that the occi- 
pito-mastoid sutures are closed in only one instance, an old animal. 
The sutures of the palate are open in all cases, except two in which the 
sagittal palatinal suture had become obliterated. The facial portions 
of the intermaxillary sutures are completely closed in only two old 
males; in four other males and one adult female these sutures had 
started to close at their lower ends. The internasal suture, which closes 
before birth in macaques (Schultz, 1937), is still entirely open in 12 
adult proboscis monkeys, partly open in 6 other adults, and com- 
pletely closed in only the remaining 8 adults. Generally speaking, it 
can be claimed that obliteration of the cranial sutures occurs com- 
paratively late in life in Nasalis in contrast to many other Old World 
monkeys and apes. Fontanellar and Wormian bones are completely 
lacking in Nasalis skulls. 

g. Pathological Conditions of the Skeleton. In conclusion it must be 
mentioned as part of the age changes in proboscis monkeys that patho- 
logical conditions become much more frec^uent as the animal gets 
older. This has already been demonstrated in regard to the dentition. 
In the skeleton the following pathological conditions have been noted : 
Among 9 infants there is one partly healed fracture of a clavicle 
(Plate 4, I). Among 7 juveniles the acromial process of one specimen 
was fractured at the base and had become fairly well healed. Among 
25 adults repaired fractures occurred in the following cases: 1. ilium 
(Plate 4, G); 2. clavicle; 3. scapula (Plate 4, F) and one rib (also long 



SCHULTZ: THE PROBOSCIS MONKEY 



309 



exostosis on shaft of r. femur); 4. corpus sterni, r. humerus and 1. 
ulna; 5. three ribs; 6. two ribs; 7. r. patella (Plate 4, H) (also exostoses 
on r. femur and tibia and corroded surfaces of r. knee-joint). In addi- 
tion disease processes were evident in two specimens: 8. arthritic 

Table 9 
The status of closure of the main cranial sutures in 26 adult proboscis 
monkeys. * = advanced attrition of the teeth. 



Specimens 


Basilary 


Coronal 


Sagittal 


Lanibdoid 


9 


open 


open 


open 


open 


1 


open 


open 


}/2 closed 


open 


1 


open 


open 


closed 


open 


1 


open 


open 


open 


closed 


1 


open 


open 


}4 closed 


closed 


1 


open 


open 


closed 


}y4 closed 


3* 


open 


closed 


closed 


closed 


1 


closed 


open 


open 


open 


2 


closed 


open 


open 


closed 


1* 


closed 


14 closed 


closed 


closed 


1* 


closed 


closed 


closed 


}/2 closed 


4* 


closed 


closed 


closed 


closed 



exostoses on many vertebrae (Plate 4, E); 9. exostoses and porous, 
irregular surface on corpus sterni and on proximal joint of 1. tibia. 
These nine specimens with manifestations of one or several skeletal 
injuries in each constitute 36 per cent of the total series of 25 skeletons 
of adult wild proboscis monkeys. Healed fractures alone occur in 28 
per cent of the adults, a frequency which is below that (33) found in 
adult orang-utans or that (36) in adult gibbons (Schultz, 1941). 



SUMMARY 

This paper represents chiefly a record of the conditions of growth 
and development in the proboscis monkey {N^asalis larvatus). Such a 
contribution will gain in significance and interest only when all its 
new data can be compared with corresponding information on the 
ontogenetic processes in representatives of many other groups of 



310 bulletin: museum of comparative zoology 

primates. Such information, based upon adequate material, is as yet 
extremely limited, so that general comparisons must unfortunately 
still be postponed. 

The series of proboscis monkeys studied consist of 51 bodies, ranging 
in age from the middle of fetal to adult life. In 41 of these specimens 
the skeletons could be investigated. An additional 2 skulls could be 
examined and the observations on the dentition have been augmented 
by the author's data for a former study on 48 specimens. 

The body weight of proboscis monkeys equals 0.45 kg. at birth, 
9.87 kg. in adult females and 20.34 kg. in adult males. The birth weight 
amounts to 4.6 per cent of the weight of fully grown females; the cor- 
responding figure for macaque is 6.7 and that for man 5.5. Proboscis 
monkeys do not seem to become pregnant until their second dentition 
is completed. All available evidence indicates that there is no breeding 
season. One of the animals contained double-ovum twins. 

The trunk of the proboscis monkey becomes more slender during 
growth, but in old males it becomes stouter again. The relative 
shoulder breadth decreases in general with age and the relative hip 
breadth changes little ontogenetically. The diameters of the pelvic 
inlet are proportionately larger in adult females than in adult males. 
The chest becomes shghtly broader in relation to the chest depth during 
growth, but it remains always proportionately narrow in comparison 
with the chests of all higher primates. The spinal column reaches 
somewhat farther into the thoracic cavity in adults than in infants, 
but at no age nearly as far as it does in the higher primates. The 
length of the tail in percentage of the trunk height increases during 
the fetal and infantile periods and decreases again thereafter. 

The rates of growth of the lengths of the limbs and the height of 
the trunk alternate with advancing age, the limbs growing faster than 
the trunk in fetal and infantile life and more slowly thereafter. In 
general the lower limbs grow somewhat faster than the upper ones 
and the distal hmb segments faster than the proximal ones. The hands 
and feet become more slender during growth and the thumb propor- 
tionately shorter. The fourth digits surpass the second ones in length 
at all ages. 

The average head diameter in its percentage relation to the trunk 
height decreases from 49 in fetuses to only 17 in adults. The cranial 
capacity averages in proboscis monkeys 70 cc. in infants, 83 cc. in 
juveniles, 85 cc. in adult females, and 102 cc. in adult males. In its 
percentage relation to body weight tliis capacity averages 7.0 in in- 
fants, 2.2 in juveniles, 0.9 in adult females, and 0.5 in adult males. 



SCHULTZ: THE PROBOSCIS MONKEY 311 

These relative cranial capacities of Nasalis do not differ significantly 
from those of langurs and macaques of corresponding weight. 

The face grows more slowly than the trunk, but more rapidly than 
the brainpart of the head. The latter grows more intensively in length 
than in breadth and height and the face increases much more in 
height than in width. The relative distance between the eyes decreases 
during the early periods of growth, but increases again later in life, 
particularly in males. The proportionate size of the outer ears reaches 
its maximum in infants and becomes smaller again in older, animals. 

Proboscis monkeys have stouter trunks, relatively longer upper ex- 
tremities and proportionately smaller outer ears than have langurs 
and macaques. 

The conditions of fetal development regarding skin color, hair, con- 
figuration of outer ears and of nose are briefly recorded. The highly 
specialized, large nose is already conspicuous in fetuses and becomes 
many times larger in adult males than in adult females. 

The deciduous teeth of proboscis monkeys erupt in the following 
order: upper il, lower i2, lower il and upper i2, lower and upper ml, 
lower c, upper c, lower m2, upper m2. This sequence is nearly the 
same as in the macaque and quite similar to that in man, but differs 
significantly from the corresponding conditions in the great apes. The 
permanent teeth make their appearance in the following order: lower 
Ml, upper Ml, resting period, lower II, upper II, lower 12, upper 12, 
lower M2, upper ]M2, lower and upper Pi and P2 (in rapid and vari- 
able succession), and finally: lower C, upper C, lower M3, upper M3 
in females and: lower M3, upper MB and lower C, upper C in males. 
Underbite occurs more often than edge-to-edge bite and overbite is 
unknown in proboscis monkeys. 

The typical vertebral formula for Nasalis is: 7 cervical, 12 thoracic, 
7 lumbar, 3 sacral, and 25 caudal vertebrae. This formula existed in 
85 per cent of the cases, indicating a comparatively high degree of 
stability. The only numerical variations found are the following: 
13 thoracic, 6 lumbar, 4 sacral, and 23, 24 and 26 caudal vertebrae. 
Transverse foramina in the seventh cervical vertebra exist bilaterally 
in only 63 per cent of the cases. The relative lengths of the cervical 
and thoracic regions become shorter during growth and the lumbar 
region increases its proportionate length, just as in the other primates 
studied so far. 

Many detailed happenings in skeletal developments have been 
recorded systematically; they are too numerous and diversified to be 
repeated here even in condensed form. The sequence of epiphyseal 



312 bulletin: museum of comparative zoology 

union is shown in a summarizing manner in table 8, where these con- 
ditions in proboscis monkeys are compared with those in some other 
primates. It appears that most of the epiphyseal lines become closed in 
an order which is approximately the same in the few primates for 
which information is available and that in Nasalis such fusion takes 
place mostly after the completion of the permanent dentition. 

At birth skeletal maturation has not progressed nearly as far in the 
proboscis monkey as in the macaque. For instance, only one ossifica- 
tion center (femur, distal) has as yet appeared in all of the epiphyses of 
the long bones in the newborn N^asaUs and the bregmatic fontanelle 
is still wide open. 

The lambdoid and sagittal sutures generally close after the comple- 
tion of the second dentition, but before the basilary and coronal 
sutures become obliterated. The coronal suture can remain open until 
old age. The internasal suture does not close until some time after 
adulthood has been attained, whereas in macaques it closes before 
birth. 

Disturbances in normal development, common among higher 
primates, are found in proboscis monkeys in comparatively few in- 
stances, of which the following are most noteworthy: congenitally 
lacking lower middle incisors, incomplete supernumerary digit, unilat- 
eral cryptorchism, divided zygomatic bones, failure to close of dorsal 
arch of atlas, of asteric fontanelles and of metopic suture. 

Pathological conditions were found among wild proboscis monkeys 
in only two young specimens (healed fractures), but in a very con- 
siderable percentage of adult animals, namely two cases with alveolar 
abscesses, seven cases with one or more healed fractures, and four 
cases with arthritic and other exostoses, etc. 



SCHULTZ: THE PROBOSCIS MONKEY 313 

BIBLIOGRAPHY 

Breschet, G. 

1845. Recherohes anatomiques et physiologiques sur la gestation des 
quadrumanes. Memoires de I'Acad. Roy. des Sciences de 1' Inst, de 
France, XIX, 401-490. 

Flower, W. H. 

1888. Einleitung in die Osteologie der Saugethiere. (Translation by H. 
Gadow; 3rd ed.) Leipzig. 

Hartman, C. G. 

1932. Studies in the reproduction of the monkey Macacus (Pithecus) 
rhesus, with special reference to menstruation and pregnancy. 
Contrib. to Embryol., XXIII, Carnegie Inst. Wash., Pub. 433, 
1-61. 

Keibel, F. 

1906. Die aussere Korperform und der Entwicklungsgrad der Organe bei 
Affenembryonen . Menschenaffen — Studien iiber Entwickelung 
und Schadelbau, 9te Liefer., Wiesbaden. 

POCOCK. R. I. 

1926. The external characters of the catarrhine monkej's and apes. 
Proceed. Zoolog. Soc. London, 1925, 1479-1579. 

ScHULTZ, A. H. 

1926. Fetal growth of man and other primates. Quart. Review of Biol., 
I, 465-521. 

1929. The technique of measuring the outer body of human fetuses and 
of primates in general. Contrib. to Embryol., XX, Carnegie Inst. 
Wa.sh., Pub. 394, 213-257. 

1930. The skeleton of the trunk and limbs of higher primates. Human 
Biol., II, 303-438. 

1933a. Die Korperproportionen der erwachsenen catarrhinen Primaten, 
mit spezieller Beriicksichtigung der Menschenaffen. Anthropol. 
Anz., X, 154-185. 

1933b. Growth and development. The Anatom}' of the Rhesus Monkey 
by C. G. Hartman and W. L. Straus, Jr., 10-27. Baltimore. 

1933c. Observations on the growth, classification and evolutionary 
specialization of gibbons and siamangs. Human Biol., V, 212- 
255 and 385-428. 

1935. Eruption and decay of the permanent teeth in primates. Amer. J. 
Phys. Anthropol., XIX, 489-581. . 



314 bulletin: museum of comparative zoology 

1937. Fetal growth and development of the Rhesus monkey. Contrib. 
to Embryol., XXVI, Carnegie Inst. Wash., Pub. 479, 71-98. 

1938a. The relative length of the regions of the spinal column in Old 
World primates. Amer. J. Phys. Anthropol., XXIV, 1-22. 

1938b. The relative weight of the testes in primates. Anatom. Rec, 
LXXII, 387-394. 

1940a. Growth and development of the chimpanzee. Contrib. to Em- 
bryol., XXVIII, Carnegie Inst. Wash., Pub. 518, 1-63. 

1940b. The size of the orbit and of the eye in primates. Amer. J. Phys. 
Anthropol., XXVI, 389-408. 

1941. Growth and development of the orang-utan. Contrib. to Embryol., 
XXIX, Carnegie Inst. Wash. Pub. 525, 57-110. 

SCHWALBE, G. 

1911. Ueber die Richtung der Haare bei den Affenembryonen. Men- 
schenaiTen^Studien liber Entwickelung und Schadelbau, lOte 
Liefer., Wiesbaden. 

WiEDERSHEIM, R. 

1901a. Beitrage zur Kenntnis der iiusseren Nase von Semnopithecus nasi- 
cns. Eine physiognomische Studie. Zeitschr. f. Morphol. u. 
Anthropol., III. 300-350. 

1901b. Nachtragliche Bemerkungen liber den Semnopithecus nasicus und 
Beitrage zur ausseren Na.se des Genus Rhinopithecus. Zeitschr. f. 
Morphol. u. Anthropol., Ill, 576-582. 



PLATE 1 



ScHULTZ — The Proboscis Monkey 



PLATE 1 

Face of an adult female proboscis monkey. This and the following two 
figures by Prof. M. Brodel. 



BULL. MUS. COMP. ZOOL. 



ScHULTz. The Proboscis Monkey. Plate 1. 




PLATE 2 



ScHULTz — The Proboscis Monkey 



PLATE 2 

Face of an adult male proboscis monkey (with little wear of teeth). 



BULL. MUS. COMP. ZOOL. 



ScHULTz. The Proboscis Monkey. Plate 2. 




PLATE 3 



ScHULTZ — The Proboscis Monkey 



PLATE 3 

Face of an old male proboscis monkey (with very marked wear of teeth). 



BULL. MUS. COMP. ZOOL. 



ScHULTZ. The Proboscis Monkey. Plate 3. 




PLATE 4 



ScHULTz — The Proboscis Monkey 



PLATE 4 

Congenital abnormalities and pathological conditions in wild proboscis 
monkeys. A = divided zygomatic bones; B = open occipito-mastoid 
fontanelles; C=open dorsal arch of atlas; D = incomplete supernumerary 
digit; E= arthritic exostoses on spinal column; F = healed fracture of r. 
scapula; G= healed fracture of 1. ihum; H = fractured patella; I = partly 
healed fracture of r. clavicle. 



BULL. MUS. COMP. ZOOL. 



ScHULTZ. The Proboscis Monkey. Plate 4. 




Bulletin of the Museum of Comparative ZoSlogy 

AT HARVARD COLLEGE 

Vol. LXXXIX, No. 7 



NOTES ON THE SPIDERS OF THE VIRGIN ISLANDS 



By Elizabeth B. Bryant 



With Three Plates 



CAMBRIDGE, MASS., U. S. A. 

PRINTED FOR THE MUSEUM 

:March, 1942 



No. 7 — N^otes on the Spiders of the Virgin Islands 
By Elizabeth B. Bryant 

The Virgin Islands, a group of small islands in the West Indies, lie 
south and east of Puerto Rico. The fifty or more islands cover ap- 
proximately two hundred and fifty square miles. The largest, St. 
Croix, St. Thomas and St. John, have been inhabited for many years, 
and at times have been intensively cultivated. 

The first comprehensive study made of the Virgin Island spiders 
was by Dr. Alexander Petrunkevitch, whose "Spiders of the Virgin 
Islands" was based on a collection made by himself in September, 
1925, and one made by Dr. Clarence R. Shoemaker in 1915. These 
two collections contained thirty-five species, of which eleven were 
described as new, eight from the Shoemaker collection. Twelve 
species had been recorded by earlier writers. Some of these have been 
found since on other islands, but four have never been recognized 
again. Of these four, three were described by C. Koch in "Die Arach- 
niden", 1836-1848, Mygale drassiformis, Evophrys vetusta and Marpissa 
incerta. The descriptions are vague and the figures are poor, so unless 
the type specimens are still in existence, the species will probably re- 
main unidentifiable. The fourth, Trochanteria ranuncida Karsch, 
1879, was in a paper of descriptions of miscellaneous spiders, mostly 
from the far east, with the locality, "Sta. Cruz." The species has never 
been found since, and it is possible that it is from another place of the 
same name. 

Aside from miscellaneous material, the Museum of Comparative 
Zoology has two collections of spiders from the Virgin Islands. The 
first was received from Dr. C. R. Wilson in 1922, while he was at the 
United States Experiment Station, St. Croix. It is a small collection, 
but each specimen has complete data. The second collection was re- 
ceived recently from Harry A. Beatty of Christiansted, St. Croix. 
The specimens are from that island. It contains a large number of 
very small spiders that are usually overlooked by casual collectors. 
The Wilson and Beatty collections contain sixty-eight species, seven- 
teen of which are new. So today the list of Virgin Island species 
numbers ninety-three, including the four uncertain ones. Although 
obviously not exhaustive, since all collections have been from the 
three largest islands only, we now have a substantial basis for a study 
of the spider fauna of the island, and its comparison with those of other 
islands of the West Indies. 



318 bulletin: museum of comparative zoology 

Of the sixty-nine species that have been examined, forty-six are 
found in Puerto Rico. This is not unexpected, as Puerto Rico is the 
nearest large island, and its spider fauna has been studied recently by 
Dr. Petrunkevitch. Some of these Puerto Rico species are, common 
house spiders and were probably introduced by commerce, as they 
are found in warm areas over the world. Others are confined to the 
New World and four are known only from these two West Indian 
island groups, though an extended survey of adjacent islands may 
show that they are more widely distributed. 

The only other island of the Lesser Antilles that has been intensively 
collected is St. Vincent, with about the same area as that of all the 
Virgin Islands combined. It is directly west of Barbadoes and much 
nearer to South America. H. H. Smith, a well known entomological 
collector, was sent there by a joint committee of the British Museum 
and the British Association for the Advancement of Science, and his 
spider collection was divided between M. Simon and the Peckhams. 
The latter described the family Salticidae and the results were pub- 
lished in the Proc. Zool. Soc. London in 1893. Fourteen species were 
recognized, of which twelve were new. Simon described the other 
families in three papers in the same publication in 1891, 1894 and 1897. 
The two authors recorded one hundred and forty-nine species in all, of 
which ninety-six were new. 

Of these one hundred and forty-nine species found on St. Vincent, 
twenty-seven have been taken on the Virgin Islands. Possibly ten 
genera are common to the two groups, each represented by an indi- 
genous species. 

About the same time that Simon was working on the St. Vincent 
fauna, he was studying collections made by himself at several places 
in Venezuela, and he found several genera common to the two places. 
This was especially noticeable in the very small spiders of the family 
Oonopidae. In a few cases, the same species was found in both places, 
and have since been found to be widespread. It is curious that none of 
these species have been found in the Virgin Island collections. Among 
other small species, Simon described a new genus, Thallumetus, 
(Dictynidae) , on males found at several places in Venezuela. Later, he 
described two more species from Chile in the same genus. Among the 
very small species in the Beatty collection from St. Croix, are males 
and females of this genus, the first time it has been recorded outside 
South America. 

In conclusion, I especially wish to express my gratitude to Mr. 
Nathan Banks for his unfailing interest and help in preparing this 



BRYANT: SPIDERS OF THE VIRGIN ISLANDS 319 

paper, and to Dr. Clarence R. Shoemaker, who so kindly made possi- 
ble the examination of certain Petrunkevitch types in his collection. 
Thanks are also due to Dr. W. J. Gertsch, who kindly sent me drawings 
of a type in the American Museum of Natural History. 

The following spiders from the Virgin Islands have been seen by me. 
The sex and the type locality follows the original description. 

Sub-order MYGALOMORPHAE 

Family THERAPHOSIDAE 

Genus Cyrtopholis Simon 1892 

Cyrtopholis pelus Chamberlin 

Cyrtopholis -pelus Chamberlin, 1917, p. 42, pi. 2, figs. 8-10. " 9 St. Thomas" 

Sub-order ARACHNOMORPHAE 

Family FILISTATIDAE 

Genus Filistata Latreille 1810 

FiLISTATA HIBERNALIS Hcutz 

Filistata hibernalis Hentz, 1842, p. 227, pi. 8, fig. 6; reprint, p. 23, pi. 2, fig. 6. 
" 9 South Carolina on the sea coast"; Petrunkevitch, 1926, p. 27. 
9 s St. Croix, (Wilson, Beatty) 

Family OONOPIDAE 

Species of the family Oonopidae are found in the dry vegetable 
detritus in warm countries. Even adults are very small. The largest 
species known is only 4.0 mm. and the smallest is less than a millimeter. 
Very few are found in temperate zones, and because they are so small, 
they are easily overlooked by the casual collector. Few species were 
known before the collection from St. Vincent was made by H. H. 
Smith. Simon proposed eight new genera for thirteen new species. 
One Old World genus was recognized, as well as one species described 
by Keyserling from Colombia. Among them, Simon found several 
that he had collected a few years before in Venezuela. Most are based 
on both male and female. 

Whether extensive collections from the various islands of the West 
Indies will show that many of the genera are endemic is uncertain. 
Dr. Petrunkevitch found five species on Puerto Rico. Two of these 



320 bulletin: museum of comparative zoology 

occur on St. Vincent, Ischnothyreus peltifer and Heteroonops spini- 
manus. Of the other three, one female was so unusual that a new genus 
was erected for it, and the two others belong to genera found on 
St. Vincent, Opopaea and Stenoonops. Eight species have been re- 
ported from Cuba, of which four are common to Puerto Rico, Oonopi- 
nus minutissimus, Heteroonops spinimanus, Ischnothyreus peltifer and 
Opopaea lutzi. For the remaining four, it was necessary to establish 
two new genera for two species, another belonged to the widespread 
genus Opopaea, and the fourth to the genus Tetrahlemma 0. P. Cam- 
bridge, 1873, based on a male from Ceylon. 

The collection made by Beatty is unusually rich in specimens of this 
family, but strangely, it has none of the species reported from Puerto 
Rico, Cuba or St. Vincent. 

Both keys to the family Oonopidae in the Histoire Naturelle des 
Araignees are full of inaccuracies and are very confusing to use. Also, 
the adjectives "high" or "low" in referring to the cephalothorax are 
misleading until enough specimens have been handled to distinguish 
between the two extremes. As Simon states in a footnote, the same 
characters are often found in the two sections of the family, (the hard 
and the soft bodies) and it is questionable if the possession or lack of a 
dorsal shield on the abdomen is of generic value, especially as it is a 
character that is sometimes wanting in the females. (Compare Cerati- 
celus, Linyjihidae) . 

The relative size and position of the eyes, possession of spines on the 
anterior legs, the elongate patellae of the anterior legs, and the 
globose or elongate coxae, seem to be characters that are constant in < 
both male and female. 

Key to the genera of Oonopidae of the Virgin Islands 

1 Abdomen with no dorsal shield 2 

Abdomen with dorsal shield 4 

2 Anterior tibiae and metatarsi with paired ventral spines, 9 ; p. I.e. 

separated from p.m.e Telchius 

Anterior tibiae and metatarsi not spined 3 

3 Cephalothorax rather low, posterior row of eyes cover less than half 

width of head, p.m.e. a broad oval Omwpinus 

Cephalothorax high, posterior row of eyes cover width of head, 
p.m.e. a narrow oval Stenoonops 

4 Anterior tibiae and metatarsi with paired spines Dysderina 

Anterior tibiae and metatarsi not spined 5 



BRYANT: SPIDERS OF THE VIRGIN ISLANDS 321 

Ventral scutum about square, followed by a very narrow scutum, 
( 9 ), anterior patellae elongate and flattened dorsally . Scaphioides 

Ventral scutum extends beyond middle 6 

Abdomen cylindrical, ventral scutum extends beyond the middle, 
no inframammillary scutum, legs long Hytanis 

Abdomen flattened, ventral scutum extends almost to spinnerets, 
small inframammillary scutum, legs short Gainasomorpha 



Genus Stenoonops Simon 1891 

Stenoonops nitens spec. nov. 

Figures 5, 10 

Male. Length, 1.3 mm. 

Cephalothorax deep yellow, granular, so that lateral margins are 
roughened, high, posterior fifth falling abruptly to margin, four-fifths 
as wide as long, anterior margin narrow, posterior margin truncate; 
eyes almost cover width of head, all eyes heavily ringed with black, 
a.l.e. round, separated by a line, largest of the six, posterior row slightly 
recurved, eyes touching because of black rings, p.m.e. elliptical, twice 
as long as wide and touching for some distance, p. I.e. round, little 
more than short diameter of p.m.e.; cJypcus convex, about equal to 
diameter of a.l.e.; mandibles small, weak and cone-shaped; labium 
wider than long, fused to sternum; maxillae, distal half much narrowed 
and inclined over labium; sternum two-thirds as wide as long, margin 
strongly lobed between coxae with a distinct lump opposite each coxa, 
tip pointed, IV coxae separated by more than a diameter, all coxae 
sub-globose; pedicel rather long; abdomen flesh-color, oval, smooth and 
shining with a slight iridescence, epigastric scutum faintly marked, 
spinnerets long, a pair of dark sacs beneath the skin separated by their 
diameter just anterior to spinnerets; legs, 4-1-2-3, varying little in 
length, no spines and rows of fine hairs, anterior patellae slightly 
elongate; palpus, pale, large, femur slender, patella and tibia sub- 
equal, little longer than diameter, cymbium and bulb united, with a 
few slender hairs on dorsal half and two very slender points at tip. 

Female. Length, 1.4 mm. 

Same as male; epigastric scutum not very distinct but the dark sacs 
anterior to spinnerets more distinct. 

Holotype cf St. Croix, (Beatty) 

Allotype 9 St. Croix, (Beatty) 

Paratypes 3 d' 1 9 St. Croix, (Beatty) 



322 bulletin: museum of comparative zoology 

Stenoonops nitens differs from the genotype in its smaller size and 
the slightly recurved posterior eye row, but it agrees with it in the 
elongate patellae of the anterior legs and the same curious palpus. 
This type of palpus has been found in species of Ischnothyreus, Sca- 
phiella, Oonopinus and Dysderina. Simon only mentions one pro- 
truding point at the tip of the bulb in Stenoonops scabiculus from 
St. Vincent and S. nitens has two. 



Genus Oonopinus Simon 1892 

OONOPINUS PRETIOSUS SpCC. nOV. 

Figures 2, 9 

Male. Length, 1.2 mm. 

Cephalothorax dull yellow, two-thirds as wide as long, smooth, rather 
low, anterior margin less than half the greatest width; eye area black, 
eyes cover less than half the width of head, a.l.e. largest of the six, 
round, separated by a radius, posterior row straight or very slightly 
recurved, p.m.e. a broad oval, touching, p.l.e. round, about equal to 
short diameter of p.m.e. and separated from them by a line; clypeus 
convex, with a few long hairs on margin, higher than diameter of 
a.l.e.; mandibles, long, cone-shaped, weak, fang long; labium about as 
wide as long, fused to sternum; maxillae with distal half narrowed, 
parallel, twice as long as labium; sternum convex, more than half as 
wide as long, tip pointed, lateral margins lobed with a distinct hairy 
point between II and III coxae, anterior coxae elongate, posterior 
sub-globose; abdomen white, cylindrical, fully twice as long as wide, 
with a few dark hairs at tip, hairs more numerous on venter, just 
anterior to spinnerets a pair of dark sacs beneath the skin; pedicel 
very short; legs not differing greatly in length; palpxis short, white, 
cymbium and bulb fused, so that palpus looks not fully developed, 
patella and tibia subequal, with many dark hairs, cymbium at base 
two-thirds as thick as long, tip pointed with embolus at tip, a white 
curved tube. 

Female. Length, 1.5 mm. 

Female the same as the male with abdomen longer, flesh-colored 
and the dark sacs on venter more distinct than in the male. 

Holotype cf St. Croix, (Beatty) 

Allotype 9 St. Croix, (Beatty) 

The genus is based on a European species, but in the same paper, 
Simon described a male and a female in the same genus from Vene- 



BRYANT: SPIDERS OF THE VIRGIN ISLANDS 323 

zuela. This species is twice as large as Ooiioyinus pretiosus, and the 
figure of the palpus shows the femur enlarged, patella greatly swollen 
and the bulb long and slender, extending at right angles from the 
cymbium. No mention is made of the sternum. 



Genus Telchius Simon 1892 

Telchius placidus spec. nov. 

Figures 3, 4 

Male. Length, 1.4 mm. 

Cephalothorax pale yellow, sides with short dark hairs that are 
thickest about posterior margin, four-fifths as wide as long, rather 
low, sloping gradually from the anterior margin to within one-quarter 
of posterior margin where it falls rapidly, anterior margin narrowed, 
no thoracic groove; eye area not covering entire width of head, a.l.e, 
sub-rotund, convex, separated by about a radius, posterior row of 
eyes recurved, p.m.e. a broad oval, touching, p. I.e. round, about equal 
to short diameter of p.m.e., narrowly separated from p.m.e. and a 
little smaller than a.l.e.; clypeus convex, about as high as diameter 
of a.l.e., many long hairs on margin; mandibles vertical, cone-shaped; 
labium fused to sternum, wider than long; maxillae about twice as 
long as labium, almost parallel; sternum heart-shaped, convex, with a 
few long hairs about margins, margin not lobed, IV coxae separated 
by less than a diameter, anterior coxae elongate, posterior sub-globose; 
abdomen oval, flesh-colored with four pairs of faint darker spots, 
venter with scanty hairs and a pair of dark marks anterior to the 
spinnerets; legs, posterior pairs longest, anterior patellae long, tibia 
and metatarsus sub-equal, no spines but many hairs, posterior pairs 
with slender, unpaired spines on tibia and metatarsus; palpus, femur 
slender, patella and tibia sub-equal, each a little longer than their 
diameter, tarsus covering about half of the bulb, bulb much swollen 
at base, extending to a point, embolus at tip, a slender dark spine. 

Female. Length, L5 mm. 

Cephalothorax, eyes and abdomen same as in male, spinnerets longer; 
legs, anterior pairs with slender spines beneath tibia and metatarsus, 
tibia, ventral, 2-2, metatarsus, ventral, 2-2, posterior pairs with 
irregular spines; epigynum a straight, transverse slit with margins 
chitinized. 

Holotype cf St. Croix, (Beatty) 

Allotype 9 St. Croix, (Beatty) 



324 Bulletin: museum of comparative zoology 

The genus is based on a female from Algeria described in a footnote in 
Simon's first paper on Spiders from Venezuela, (Ann. Soc. Ent. France, 
1892, 61, p. 446.) In the text, a second species, also a female is de- 
scribed. The genus differs from Oonops by the almost parallel maxillae, 
slender spines on the anterior legs and the elongate patellae of the 
anterior pairs. The male of this species has no spines on the anterior 
tibiae but the female has two pairs. The pair apparently belong 
together, as the eyes are the same, both have short hairs on the 
cephalothorax and the abdomen. 



Genus Dysderina Simon 1891 

Dysderina antillana spec. nov. 

Figures 1, 7 

Male. Length, 1.5 mm., ceph. 0.7 mm., abd. 0.7 mm. 

Cephalothorax olive-brown, darker about lateral margins, nine- 
tenths as wide as long, head narrowed, carapace very high, rising from 
eye area to one-fifth from posterior margin where it falls abruptly; 
eyes cover fully one-half the width of the head, a. I.e. largest of the six, 
round, narrowly separated, posterior row straight, only slightly longer 
than anterior row, p.m.e. a broad ellipse, touching for some distance, 
p. I.e. round, smaller than a. I.e. and touching p.m.e.; chjpeus equals 
about a radius of a. I.e., slightly convex; viandiblcs vertical, cone-shaped, 
pale, with a small dark hook that projects forward at base of the fang, 
fang groove short; labium pale, fused to sternum, one-half as long as 
wide, sides parallel; maxillae slightly inclined, distal half very narrow 
and parallel; sternum as wide as long, widest between II and III coxae, 
squarely truncate between IV coxae, convex, smooth, no hairs, IV 
coxae separated by a diameter and a half; abdomen almost two-thirds 
as wide as long, cylindrical, an olive-brown scutum covers two-thirds 
of dorsum, sides iridescent white with a few hairs, ventral scutum 
only slightly chitinized, extends from pedicel to within one-third of 
the spinnerets for almost the entire width of the abdomen, tip truncate, 
each side of the spinnerets a pair of dark sacs beneath the skin, spin- 
nerets long; legs, III right and IV left missing, all coxae sub-globose, 
legs slender, pale, anterior patellae not elongate, spines, I pair, femur, 
1-1 prolateral near tip, tibia, ventral, 2-2-2-2, long and overlapping, 
metatarsus, ventral, 2-2, long and overlapping, posterior pairs spine- 
less; palpus, small, slender and dark. 

Holotype cf St. Croix, (Beatty) 



BRYANT: SPIDERS OF THE VIRGIN ISLANDS 325 

Simon based the genus Dysderina on three species found at St. Vin- 
cent, two that were new and a third that was originally found by Key- 
serling in Colombia and described in the genus Oonops. A year later, 
Simon described another species from Venezuela. Today, the genus is 
recognized as widely distributed, and has been divided in two sections, 
the first, with the posterior row of eyes straight, eyes large and con- 
tiguous and the a.l.e. separated by less than their diameter; the second 
section, with the posterior row of eyes procurved, p.l.e. a little sepa- 
rated from p.m.e. and a.l.e. separated by more than their diameter. 
Dysderina antillana belongs to the first section. It differs from its rela- 
tives in the small size and the palpus. As far as known, no species has 
been found with a hook on the mandibles above the base of the fang 
but this could be easily overlooked in so small a spider. However, it 
is very distinct under high magnification because of the dark color. 



Genus Gamasomorpha Karsch 1881 

Gamasomorpha perplexa spec. nov. 

Figures 11, 12, 18 

Male, Length, 1.5 mm., ceph. 0.8 mm., abd. 0.7 mm. 

Cephalothorax bright brown, two-thirds as wide as long, (3 : 4.5), 
sparsely punctate, anterior margin narrowed, posterior margin slightly 
concave, no thoracic groove, carapace slopes gradually from anterior 
margin to within one-tenth of posterior margin, where it falls rapidly 
in a concave slope; eyes not covering entire anterior margin, a.l.e. 
largest of the six, separated by a little more than a radius, posterior 
row almost straight or slightly recurved, p.m.e. a broad ellipse, touch- 
ing, p.l.e. round, less than a short diameter of p.m.e. and touching 
p.m.e. ; clypen^ narrow, less than a radius of a.l.e. ; mandibles vertical, 
small and cone-shaped; labium almost triangular, fused to sternum; 
maxillae about twice as long as labium, inclined, narrow and lateral 
margins almost parallel; stermnn triangular, more than two-thirds as 
wide as long, (2 : 2.5), convex and smooth, lateral margins distinctly 
lobed and area between elevated, tip truncate, IV coxae separated by 
more than a diameter; pedicel short; abdomen oval, brown, scutum 
covers entire dorsum, flat, punctate, probably in life a small hair from 
each pit as a few remain above spinnerets, venter almost completely 
covered by a scutum which meets the dorsal scutum at base and shows 
no opening for lung slits, a small infra-mammillary scutum on ventral 
side of spinnerets; legs short, pale brown, no spines, anterior coxae 



326 bulletin: museum of comparative zoology 

elongate, posterior coxae globose, femora only slightly enlarged and 
laterally compressed; palpus, very dark brown, small and rather 
short, femur not enlarged, tibia globose, patella a little longer than 
tibia, tarsus with a dorsal row of seven clavate hairs, tarsus longer than 
tibia plus patella, palpal organ extends beyond cymbium almost the 
length of the tarsus, embolus at tip. 

Female. Length, 1.5 mm. 

Female the same size as the male with the same scuta and eyes as in 
the male; ventral scutum apparently divided at fold and the opening 
of the epigynum a small round depression in a pale area just posterior. 

Holotype cf St. Croix, (Beatty) 

Allotype 9 St. Croix, (Beatty) 

Paratypes 2 9 St. Croix, (Beatty) 

According to the definition of the genus in the Histoire Naturelle 
des Araignees, 1893, 1, p. 301, there is great variation in the size of the 
eyes of the posterior row, height of the clypeus, but the coxae are all 
globose. This species differs in having the anterior coxae elongate and 
a greater discrepancy in the size of the eyes of the posterior row than 
usual, but the legs are short and there is an infra-m^mmillary scutum. 



Genus Hytanis Simon 1892 

Hytanis pusilla spec. nov. 

Figures 13, 14 

Female. Length, 1.5 mm., ceph. 0.5 mm., abd. 1.6 mm. 

Cephalothorax golden-brown, with a few long hairs, irregularly 
placed, almost as wide as long, (4.2 : 4.5), convex, but not very high, 
slopes gradually from anterior margin to opposite III pair of legs when 
it falls rapidly to posterior margin, anterior margin about half the 
greatest width, no thoracic groove; eyes, a. I.e. largest of the six, sepa- 
rated by two-thirds of a diameter, posterior row slightly recurved, 
longer than anterior row, p.m.e. a broad ellipse, touching, p. I.e. sepa- 
rated from p.m.e. by a little more than a line, smaller than a. I.e.; 
clypeus equals about a diameter of a. I.e., convex; mandibles long, cone- 
shaped, fang groove short; labium wider than long, fused to sternum; 
maxillae short, about one and a half times as long as labium, slightly 
inclined; sternum as wide as long, anteriorly truncate, lateral margins 
almost parallel, not emarginate between coxae, strongly convex, with 
a few long hairs, IV coxae separated by more than a diameter; abdomen 
more than twice as long as wide, convex, dorsum covered by a reddish- 



BRYANT: SPIDERS OF THE VIRGIN ISLANDS 327 

brown scutum, sparsely punctate with a long colorless hair from each 
depression, a faint pattern can be traced on the scutum with a median 
pale stripe and darker spots each side, ventral scutum from pedicel 
to fold, followed by a paler scutum which passes the middle of venter, 
no scutum about spinnerets; legs, IV leg longest, all coxae sub-globose, 
femora only slightly compressed, patellae not greatly elongate, all 
legs with many long colorless hairs or bristles, no spines. 

Holotype 9 St. Croix, (Beatty) 

The genus Hytanis was based by Simon on a female from Puerto 
Cabello, Venezuela. Both the generic and specific descriptions are very 
brief. It is compared with Gamasomorpha, but no mention is made of 
the ventral scutum. The elongate abdomen is not common in the 
Oonopidae. It cannot be placed in the genus Gamasomorpha because 
the ventral scutum is short and the infra-mammillary scutum is lacking; 
also the legs are long and slender. 

Genus Scaphioides genus nov. 

Cephalothorax moderately high; eye area covering about half the 
anterior margin, a. I.e. largest of the six, posterior row straight, p.m.e. 
an elongate ellipse and touching for some distance, separated from 
p.l.e. by little more than a line; all coxae sub-globose; sternum convex 
and lateral margins carried between coxae; abdomen, (female), with 
no dorsal scutum, epigastric scutum about square, from pedicel to 
epigastric fold, followed by a very narrow scutum which is little more 
than a line; legs spineless, anterior patellae elongate and dorsally flat- 
tened; male unknown. 

Genotype Scaphioides reducta spec. nov. 

This genus differs from Scaphiella Simon in the abbreviated ventral 
scutum in the female and the greatly elongate p.m.e.; it agrees with 
that genus in the straight posterior eye row and the elongate anterior 
patellae. 

Scaphioides reducta spec. nov. 

Figures 6, 8 

Female. Length, 1.6 mm., ceph. 0.5 mm., abd. 1.0 mm. 

Cephalothorax golden-brown, sides granular, almost three-quarters 

as wide as long, anterior margin much narrowed, posterior margin 

convex, moderately high, slopes gradually from anterior margin to 

wathin one-fifth of posterior margin when it falls rapidly in a convex 

slope; eye area covering about half the width of the head, a. I.e. largest 



328 bulletin: museum of comparative zoology 

of the six, separated by a line, posterior row straight, longer than an- 
terior row, p.m.e. a narrow ellipse, touching for some distance, p. I.e. 
round, little less than long diameter of p.m.e. and separated from p.m.e. 
by a line; clypeus convex, a little higher than diameter of a. I.e.; man- 
dibles small, vertical; labium as long as wide, tip much narrowed; 
maxillae inclined over labium, but tips not touching, distal half much 
narrowed; sternum strongly convex, granular, lateral margins carried 
between coxae, IV coxae separated by more than a diameter, all coxae 
sub-globose; abdomen flesh-color, oval, convex, sparsely covered with 
short hairs, venter paler, with short hairs, a square scutum from 
pedicel to fold, followed by a scutum that is little more than a line, 
spinnerets short, each side, beneath the skin an irregular dark sac; 
legs moderately long, anterior patellae elongate and dorsally flattened, 
I patella three-quarters as long as I tibia, no spines but rows of short, 
colorless hairs; epigynuvi a long, transverse slit between epigastric 
scutum and the very narrow ventral scutum. 

Holotype 9 St. Croix, (Beatty) 

Paratype 9 St. Croix, (Beatty) 



Family CAPONIIDAE 

Genus Caponina Simon 1891 

Caponina blanda spec. nov. 

Figure 17 

Male. Length, 4.6 mm., ceph. 2.1 mm., abd. 2.5 mm. 

Cephalothorax bright orange, smooth, no hairs, oval, convex, more 
than two-thirds as wide as long, (4.5 : 6.0), anterior margin rounded 
and narrow, no thoracic groove; eyes on a black spot, two, round, 
separated by about a radius; clypeus at least three times the diameter 
of an eye, convex and sloping; mandibles vertical, shining, cone-shaped, 
fang groove short; labium longer than wide, fused to sternum; maxillae 
only little longer than labium, curved, so that tips almost meet, palpi 
inserted above the middle, distal half very little narrowed; sternum 
deep orange, shining, oval, about half as wide as long, lateral margins 
indentate, flat, IV coxae separated by more than a diameter; abdomen 
oval, dull green, scantily covered with short colorless hairs, venter little 
paler than dorsum; legs little paler than cephalothorax, 4-1-2-3, little 
difference in length, no spines and few hairs, I coxa longer than IV 
coxa. III coxa much smaller than others; palpus as long as cephalo- 



BRYANT: SPIDERS OF THE VIRGIN ISLANDS 329 

thorax, pale yellow, femur long and slender, with no rasping spur, tibia 
longer than patella, eymbium as long as patella plus tibia, bulb globose 
and protruding from the cavity, embolus a pale, stout tube which 
leaves the bulb at base and becomes free before the middle. 

Holotype cf St. Croix, (Beatty) 

The genus Caponina was based by Simon on a female from St. Vin- 
cent. Caponina blanda differs in several points from the generic de- 
scription, but these differences may be sexual. Caponina testacea, the 
genotype, is described as having the clypeus scarcely twice as high as 
the diameter of an eye and subvertical, the sternum is convex. In 
Caponina blanda, the clypeus is fully three times the diameter of an 
eye, and it is convex and sloping, the sternum is flat. The genus in- 
cludes very few species, and the sexes are rarely found together. 



Family OECOBIIDAE 

Genus Oecobius Lucas 1845 

Oecobius benneri Petrunkevitch 

Oecobius benneri Petrunkevitch, 1929, p. 75, figs. 64-66. "9 Puerto Rio; 

Rio Piedras" 
Oecobius parietalis Petrunkevitch, 1926, nee Hentz, 1850. 
9 St. Croix, (Beatty) 

Family LYCOSIDAE 
Genus Lycosa Latreille 1S04 

Key to Females 

1. Cephalothorax with a narrow median pale stripe extending to 

anterior eye row ^ atlantica 

Cephalothorax with median pale stripe as wide as space between 
p. 1. e 2 

2. Sternum with a dark stripe 3 

Sternum pale 4 

3. Venter almost covered with a dark spot sancii-vincenti 

Venter pale sancti thomasi 

4. Anterior tibiae with small weak spines, spider about 20.0 mm. 

long reduda spec. nov. 

Anterior tibiae with normal spines, spider about 12.0-13.0 mm. 
long subtilis spec. nov. 



380 bulletin: museum of comparative zoology 

Three species of Lycosa with a dark stripe on the sternum have been 
described from the West Indies. Two of these have a dark spot on the 
venter and all have an angulate tip to the basal spear-mark on the 
abdomen. 

Lycosa sandi-vincenti Simon was described from both male and 
female, but no figures are given of the palpus and the epigynum. This 
species has three pairs of ventral spines on the first tibia and two 
prolateral spines. It is possible that Lycosa yaucensis Petr. from Puerto 
Rico is a synonym. This is described with one prolateral spine on the 
first tibia. 

Lycosa sancti thomasi Petr., 1926, known only from the type, is 
about the same size, but all the dark marks are very faint, and the dark 
basal spear-mark on the abdomen fades about the middle, so that the 
tip cannot be seen. The prolateral spines on the first tibia are confusing 
as one leg has one spine and the other has two. 

From the descriptions of all three species, the epigynum has the 
median septum slightly angulate about the middle. 

Lycosa atlantica Marx 

Lijcosa atlantica Marx, 1889, p. 100, pi. 4, fig. 4. " 9 Bermuda" 
5 9 St. Croix, (Beatty) 

Lycosa reducta spec. nov. 
Figure 25 

Female. Length, 19.0 mm., ceph. 9.5 mm., abd. 8.5 mm., I tib. 
5.0 mm. 

Cephalothorax mahogany -brown, median pale stripe from p. I.e., 
margins gradually converging to posterior margin, in anterior half a 
pair of narrow dark stripes that do not reach the thoracic groove, a 
very narrow median stripe of white hairs from p. I.e. to anterior eye 
row, lateral pale stripes not as wide as dark stripe, with irregular margins, 
sides below eyes covered with fine white hairs; eyes, anterior row not 
as wide as second row, eyes equidistant, a.m.e. larger than a. I.e., eyes 
of second row separated by less than a diameter; viandihles dark brown, 
geniculate, three teeth on each margin; abdomen with a median spear 
mark that does not reach middle, tip forked, surrounded by a pale 
stripe, posterior half mottled and cross bars vaguely indicated, sides 
grayish-brown, entire dorsum with many long bristles, venter pale; 
legs with the two terminal joints darker, posterior femora with broken 



BRYANT: SPIDERS OF THE VIRGIN ISLANDS 331 

dark rings on dorsal side, spines, no spines on anterior patellae, poste- 
rior patellae with 1 prolateral and 1 retrolateral, I pair, tibia, dorsal, 
0, prolateral, 2, retrolateral, 0, ventral, 2-2-2, all spines very slender 
and weak, less than one-half the diameter of the joint, area between 
ventral spines thickly covered with short, white hairs, metatarsus, 
dorsal, 0, lateral, 0, ventral, 2-2, in a thick scopula, so that spines are 
hidden, basal and median, both pairs very small, II pair, spines the 
same as on I pair, but longer and scopula not as heavy. III and IV 
tibiae with a dorsal, basal spine, all spines on posterior pairs are long 
and heavy; epigynuvi very similar to Lycosa subtilis spec, no v. (cf. 
posiea) . 

Holotype 9 St. Croix, (Beatty) 

This species is very near Lycosa subtilis but is much larger. The legs 
in proportion are shorter and stouter, the spines on the anterior tibiae 
and metatarsi are small and weak, and the anterior metatarsi have a 
thick scopula. Lycosa aussercri (Keys.) from Colombia is about this 
size, but the spines of the anterior tibiae and metatarsi are described 
as small and weak. Banks identified three specimens from Culebra 
and Vieques Islands as this species. Petrunkevitch mentions them in 
"The Spiders of Porto Rico", 1929, p. 91 and regarded them as prob- 
ably Lycosa fusca (Keys.), 1876, described from Cuba. They are prob- 
ably not Lycosa aussercri from Colombia, but it is equally questionable 
if they are fu^ca. The two islands lie between Puerto Rico and St. 
Croix, and they may prove to be reducta. 

Lycosa sancti thomasi Petrunkevitch 
Lycosa sancti thomasi Petrunkevitch, 1926, p. 67, fig. 22. " 9 St. Thomas" 

Through the kindness of Dr. Shoemaker, I have been able to exam- 
ine the unique type specimen. 

The dark markings on the cephalothorax and abdomen are rather 
obscure; the median pale stripe on the cephalothorax is very little con- 
stricted anterior to the groove and the narrow pale lateral stripes are 
only faintly indicated; the abdomen has a dark basal mark that dis- 
appears about the middle, the dark median spot on the sternum is very 
faint and the entire venter is pale. The specimen has not oviposited. 

The spines on the legs are confusing. The anterior patellae have no 
spines, the first tibia has three pairs of ventral spines, and the left leg one 
small prolateral spine beyond the middle ventral pair, the right leg 
has two prolateral spines, one beyond the middle pair and one much 
nearer the base. It is possible that one leg has been replaced but as 



332 bulletin: museum of comparative zoology 

there is no difference in length or size, it is impossible to say which leg 
has been renewed. 

Lycosa sancti-vincenti Simon 

Lycosa sancti-vincenti Simon, 1S97, p. SSS. "d" 9 St. Vincent" 

It is not impossible that Dr. Petrunkevitch had a small, pale speci- 
men of this species from Puerto Rico, when he described Lycosa yau- 
censis. The Simon type is 12.0 mm. long, the venter is black and the 
sternum is black with a yellow margin; spines, I pair, patella, prolateral 
1, tibia, ventral, 2-2-2, prolateral, 2 small spines, metatarsus, ventral, 
2-2, apical whorl of 3 spines; epigynum is much longer than wide, 
subparallel and the margins reddish. Four specimens from St. Croix, 
collected by Dr. Wilson in 1922, and one from Antigua, collected in 
1918, answer this description. A single specimen from St. Croix, col- 
lected by Beatty is smaller, (9.5mm.) and agrees very well with the 
description of Lycosa yavcensis Petr. The smaller specimen has ir- 
regular brown marks on femora and tibiae and the first tibia has but 
one prolateral spine. The figure of the epigynum has the median sep- 
tum broad, slightly wider and angulate about the middle. All speci- 
mens seen have the same epigynum with the margin of the septum 
heavily chitinized. 

4 9 St. Croix, (Wilson) 

19 St. Croix, (Beatty) 

Lycosa subtilis spec. nov. 
Figures 16, 19 

Female. Length, 12.5 mm., ceph. 6.6 mm., abd. 6.5 ram., I tib. 
3.1 mm. 

Cephalothorax with median pale stripe from p.l.e., narrowing before 
the thoracic groove, anterior half of pale stripe with narrow curved 
dark stripes ending opposite I coxae, lateral pale stripes rather broad, 
dark stripes covered with dark, short hairs; eyes, anterior row shorter 
than second row, eyes equidistant, a.m.e. separated by a radius, larger 
than a.l.e., eyes of second row separated by less than a diameter, scat- 
tered white hairs and longer bristles in eye area; inandibles mahogany- 
brown, geniculate, with many coarse dark bristles and shorter white 
hairs, fang groove oblique, thick scopula of long hairs at base of fang 
on superior margin, three teeth on each margin; labium longer than 
wide, lateral margins deeply notched at basal half, tip rebordered; 



BRYANT: SPIDERS OF THE VIRGIN ISLANDS 333 

maxillae twice as long as labium; abdomen oval, dark basal mark ex- 
tends to middle with tip truncate and each lateral corner extended in a 
a diagonal line that connects with lateral stripe, pale lateral stripes 
meet at base and fade before middle, distal half dark, with vague cross 
bars and four or five pairs of widely separated pale spots, venter pale 
with no marks; legs pale, distal joints darker and on posterior pairs 
very faint dark rings, most distinct on dorsal side, spines, no spines on 
anterior patellae, posterior pairs, 1 prolateral, 1 retrolateral, I pair, 
tibia, dorsal, 0, prolateral, 1, retrolateral, 0, ventral, 2 distal, 2-2, 
both pairs rather slender and the retrolateral row little more than 
bristles, area between with a thick covering of short hairs, but not as 
heavy as a scopula, metatarsus, dorsal, 0, lateral, 0, ventral, 2 basal, 
2 median, 1 distal, II pair, spines same as on I pair but ventral hairs 
not as thick on tibia, III and IV tibiae with basal, dorsal spine; 
epigynum a median narrow septum, sides almost parallel and longer 
than cross piece. 

Male. Length, 11.3 mm., ceph. 5.6 mm., abd. 5.5 mm. 

Cephalothorax with median pale stripe which starts between p.m.e. 
and widens to width of p. I.e., gradually narrows to posterior margin, 
the pair of dark stripes in pale area not as long as in female, and fol- 
lowed on lateral margin by a pair of distinct dots, these can be traced 
in the female, marginal dark stripe narrower than in female, line of 
bristles below second and third eye row very distinct; mandibles, 
labium and sternum as in female; abdomen very distinctly marked, 
lateral pale stripes surround median dark spear mark and continue to 
spinnerets with posterior half crossed by broken chevrons, the pale 
spots reduced to two or three pairs, venter pale; legs pale, distal joints 
darker, interrupted dark rings can be traced on posterior pairs, spines, 
all patellae with 1 prolateral and 1 retrolateral spine, dorsal, basal 
spine on posterior tibiae, I pair, tibia, prolateral, 1, retrolateral, 2, ven- 
tral, 2-2-2, all spines large and heavy, metatarsus, dorsal, 0, pro- 
lateral, 2, retrolateral, 2, ventral, 2-2, 1 median apical spine; yalpus 
as long as cephalothorax, slender, pale, tibia almost twice as long as 
patella and three-quarters as long as cymbium, tip of cymbium very 
narrow, the parts of the palpus very much like Lycosa pratensis Emer- 
ton, but larger. 

Holotype cf St. Croix, (Beatty) 

Allotype 9 St. Croix, (Beatty) 

Paratypes 5 9 St. Croix, (Beatty) 

This species probably belongs in the same section of the genus as 
Lycosa pratensis Emerton, but it is much larger. The spines on the 



334 bulletin: museum of comparative zoology 

legs are not the same in the male and female but the two undoubtedly 
belong together, as they have the same marks on the cephalothorax 
and abdomen. The female of Lycosa pratmsis has the retrolateral row 
of ventral spines on the first tibia much reduced. 

Genus Pardosa C. Koch 1848 

Pardosa portoricensis Banks 

Pardosa portoricensis Banks, 1901, p. 224, pi. 15, figs. 2, 3. "1 9 Porto Rico; 
San Juan" 

2 9 St. Croix, (Beatty) 



Family OXYOPIDAE 

Genus Oxyopes Latreille 1804 

OxYOPES SALTicus Hentz 

Oxyopes salticus Hentz, 1845, p. 196, pi. 16, fig. 10; reprint, p. 47, pi. 6, fig. 10. 
"North Carolina, Alabama" 

2 cf 2 9 St. Croix, (Beatty) 



Family SICARIIDAE 

Genus Scytodes Latreille 1804 

ScYTODES FUSCA Walckcnaer 

Scytodes fusca Walckenaer, 1837, 1, p. 272 'V 9 Cayenne"; Petrunkevitch, 
1926, p. 29. 

5 9 St. Croix. U. S. Exper. Stat., (Wilson.) 
2 9 St. Croix, (Beatty) 

Scytodes hebraica Simon 

Scytodes hebraica Simon, 1891, p. 568, pi. 42, fig. 14. " 9 St. Vincent" 
9 St. Croix, 6 April 1922, (Wilson) 

Scytodes lineatipes Tacz 

Scytodes lineatipes Taczanewski, 1873, p. 107. "2 9 Cayenne et une des lies 
du Salut" 

immat. 9 St. Croix, June 1922, (Wilson) 



BRYANT: SPIDERS OF THE VIRGIN ISLANDS 335 



SCYTODES LONGIPES LucaS 



Scytodes longipes Lucas, 1845, p. 71, pi. 1, fig. 2. "c? Mexique" 
9 St. Croix, 24 April 1922, (Wilson) 
immat. 9 St. Croix, (Beatty) 



Family PHOLCIDAE 

Genus Artema Walckenaer 1837 

Artema ATLANTA Walckenaer 

Artema atlanta Walckenaer, 1837, 1, p. 656, "cf 9 Amer. merid 

du Brazil"; Petrunkevitch, 1926, p. 27. 
9 St. Croix, (Beatty) 

Genus Modisimus Simon 1893 

MoDisiMus MONTANUS Petr. 

Modisimus rnontanusV&trajakevitch, 1929, p. 131, figs. 120-126. "cf 9 Porto 
Rico" 

Petrunkevitch described the female as 3.5 mm. long. Seven females 
and a male were found under rocks between Lares and Yauque and 
another female at Guainabo. Three females were found by Beatty 
that probably belong to this species, although they are about half the 
size, (1.5 mm.). They have the same arrangement of eyes, with a dark 
stripe from the eye turret to margin of the clypeus and the same 
epigynum. 

3 9 St. Croix, (Beatty). 

Genus Physocyclus Simon 1893 

Physocyclus globosus (Tacz.) 

Pholcus globosus Taczanowski, 1873, p. 105. "3 9 Cayenne" 
9 St. Croix, 27 March 1922, (Wilson) 

Genus Smeringopus Simon 1890 

Smeringopus elongatus (Vinson) 

Pholcus elongatus Vinson, 1864, p. 135. " 9 , ile de la Reunion" 
cf 9 St. Croix, March, June 1922, (Wilson) 



336 bulletin: museum of comparative zoology 

Family DICTYNIDAE 

Genus Thallumetus Simon 1892 

Thallumetus parvulus spec. nov. 

Figures 20, 22, 23 

Male. Length, 1.0 mm. 

Cephalothorax brown, posterior portion of head with a few gray 
stripes which meet at beginning of thoracic portion, head very high, a 
median line of curved bristles from anterior eye row to thoracic por- 
tion, radial striae shaded; eyes, anterior row straight, a.m.e. small, 
separated by a diameter, almost touching a. I.e., a. I.e. about twice the 
diameter of a.m.e., posterior row slightly procurved, almost straight, 
little longer than anterior row, p.m.e. separated by little less than a 
diameter, subequal with a. I.e. and separated from p. I.e. by about a 
radius, lateral eyes touching; quadrangle wider behind than in front 
and as high as wide; clypeus higher than diameter of a.l.e. with a few 
long bristles in middle directed downward; mandibles long, swollen at 
base, median margin excavate about middle; labium longer than wide; 
sternum pale, convex, widest between II coxae and pointed between 
IV coxae, IV coxae separated by a diameter; abdomen oval, dorsum 
flattened, covered with long hairs, a median basal dark stripe which 
does not reach the middle, with pale lateral stripes almost as wide 
as dark stripe, these meet and continue to near the spinnerets, lateral 
margins of pale stripe irregular, venter pale, spinnerets, basal pair 
separated by a scant diameter; legs, pale, varying little in length, cov- 
ered with hairs, no spines; palpus pale, shorter than cephalothorax, 
femur greatly swollen, so that it is more than half as wide as long, 
with a dark ventral chitinized carina near tip, this evidently rubs on a 
flattened lobe on ventral side of tibia, patella swollen on dorsal side, 
tibia very short and almost covered by patella, with a bifid prolateral 
process much chitinized, cymbium prolonged in a point, bulb circular 
and a little convex, embolus a stout black spine from near the tip which 
follows the contour of the cavity to near base. 

Female. Length, 1.3 mm. 

Cephalothorax same as in the male but the median row of bristles 
does not extend beyond the posterior eye row; eyes and clypeus same 
as in male; ma?idibles, median margin not as much excavate; sternum 
and legs the same; abdomen larger than in male, covered with short 
hairs that in some lights are iridescent, lateral pale stripes meet poste- 
rior to the dark basal spear mark and end abruptly with a transverse 



BRYANT: SPIDERS OF THE VIRGIN ISLANDS 337 

line about one third from the tip, venter pale, cribellum not divided, 
calamistrum a single row of 10-12 curved hairs, extends from near 
base to very near tip of IV metatarsus; spinnerets separated, basal 
pair longest and separated by about two diameters, median pair very 
small and touching, superior pair about two-thirds as long as basal and 
separated by more than a diameter; epigynum very faintly marked, 
apparently two oval openings that are separated by at least their 
long diameter. 

Holotype cf St. Croix, (Beatty) 

Allotype 9 St. Croix, (Beatty) 

Paratype cf 2 9 St. Croix, (Beatty) 

The genus Thalluvietus was described in the Histoire Naturelle des 
Araignees, 1, p. 241 which was published in October, 1892. It is based 
on males found in three localities in Venezuela. The genotype species 
was described in the Ann. Soc. Ent. France, 61, p. 434, pi. 9, fig. 7, 
which was issued in April 1893, six months after the description of 
the genus. Thallumetus is separated from Didyna in both sexes, by 
the slightly procurved posterior eye row and in the female by the widely 
separated spinnerets, in the male these are not as widely separated. 
The male palpus has the femur much enlarged. The genotype is much 
larger than the St. Croix species and differs in a few minor points. 
Thallumetus salax has the eyes of the anterior row equal and the palpus 
is figured with the patella slender, while in Thallumetus parvulus the 
eyes of the anterior row are very unequal and the patella of the palpus 
is greatly swollen on the dorsal side. 



Family THERIDIIDAE 
Genus Lithyphantes Thorell 1870 

LiTHYPHANTES SEPTEM]\L\CULATUS Keys. 

Lithyphantes septemmaculatus KeyserKng, 1884, p. 141, pi. 6, fig. 88. " 9 
Denver, Colorado; Enterprise, Fla." 
9 s St. Croix, (Beatty) 

Genus Conopistha Karsch 1881 

Three species of the genus Conopistha have been found on St. Croix. 
The males are easily separated by the anterior horn and the palpus. 
I do not agree with Dr. Petrunkevitch in his description of the species 
found on Puerto Rico, that the structure of the male palpus is not a 



338 bulletin: museum of comparative zoology 

safe character to use in the identification of species, (1930, p. 179). 
The genus is easily separated, in the males, into two sections by the 
shape of the cymbium, the first with the distal end of the palpus pro- 
duced in a prolateral lobe, and the second with the distal end con- 
tracted in a rounded tip. 

Taczanowski was the first author to recognize the genus in the 
New World and in "Les Araneides de la Guyane fran^aise", 1872, he 
describes several species. Of these, probably nephilae has been re- 
ported most often. Unfortunately, this species was not figured until 
1880, when O. P. Cambridge wrote his paper on the genus. It was 
evident that he never saw the types, and he confused two species. In 
1884, Keyserling, who had the Taczanowski types, corrected the error 
made by Cambridge in "Die Spinnen Amerikas", but for some strange 
reason this correction has never been followed. Keyserling also sug- 
gested that lugens and jucunda, both described by Cambridge from 
the Amazons, are the females of nephilae Tacz. In the text, Cambridge 
suggests that lugens may be the female of conciiina which is con- 
sidered as a synonym of nephilae Tacz. 

In 1891, Keyserling in "Die Spinnen Amerikas, Brasil," p. 215, re- 
named the species that Cambridge had misnamed nephilae as cam- 
bridgei. This is the species that is so widely distributed in northern 
South America, most of the islands of the West Indies and the southern 
part of the United States. 

The females of the genus are not easy to place. The epigynum is 
often filled with a dark brown substance, so that the openings are im- 
possible to see and the dark color on the venter may extend on the 
sides or even to the dorsum, so that the color pattern is of little value. 



CoNOPiSTHA cambridgei (Keys.) 
Figures 21, 24 

Arc/yrodes iiepJdlae, O. P. Cambridge, 1880, p. 324, nee Taczanowski, 1872, 
"cf 9 Amazons"; Petrunkevitch, 1929, p. 179, figs. 19-22, nee Taczan- 
owski, 1872. 

Argyrodes cambridgei Keyserling, 1891, p. 215. 

Male. Length, 2.0 mm., ceph. 1.0 mm., abd. 1.0 mm. 

Cephalothorax. yellowish-brown, flat, two-thirds as wide as long, 
thoracic groove transverse, anterior horn very slender, starts midway 
between margin of the clypeus and the eyes, with the ventral surface 
concave, best seen in a lateral view, tip a rounded knob projects out- 



BRYANT: SPIDERS OF THE VIRGIN ISLANDS 339 

ward, covered with hairs and is as narrow as space between a.m.e.; 
eyes median eyes project forward on a lobe, p.m.e. separated by more 
than a diameter, lateral eyes touching, about twice as far from space 
between horn and eye lobe, as a.l.e. plus p. I.e.; clypeus vertical from 
margin to base of horn; labium fused to sternum; sternum dark brown, 
triangular, widest between I pair of coxae, IV coxae separated by 
almost a diameter, lateral margins carried between coxae; abdomen 
silvery, often with a median dark line from base to near tip, posterior 
part of abdomen high, so that height over spinnerets equals length, 
tip produced upwards, rather than backwards, tip dark, basal part of 
abdomen a narrow chitinized ring, with anterior margin roughened, 
can be seen from dorsal side, venter almost black, with a pair of 
silvery spots anterior to spinnerets, margin between black venter 
and silvery sides scalloped; palpus pale, as long as cephalothorax, 
patella swollen dorsally, cymbium with tip widened, with a large hairy 
prolateral lobe at end, this lobe rounded and not chitinized as in 
Conopistha nephilae, best seen in the figure. 

Female. Length, 2.0 mm., ceph. 1.0 mm., abd. 1.0 mm. long, 
1.3 mm. high. 

Abdomen with the same markings as in the male but often higher. 

cf s 9 s St. Croix, (Beatty) 

Conopistha cancellatus (Hentz) 

Theridion cancellatum Hentz, 1850, p. 278, pi. 19, figs. 17, 18; reprint, p. 149, 
pi. 16, figs. 17, 18. 

cf 9 St. Croix, (Beatty) 

Conopistha nephilae (Tacz.) 
Figures 15, 26 

Argyrodes nephilae Taczanowski, 1872, p. 114. " d^ 9 Cayenne, des lies du 
Salut et de Saint Laurent de Alaroni;" Keyserling, 1884, p. 184, pi. 8, 
fig. 110. 
Argyrodes coucinna O. P. Cambridge, 1880, p. 322, pi. 38, fig. 2. " ? Amazons" 
Argyrodes lugens id., ibid., p. 327, pi. 38, figs. 2a', b', c'. " 9 Amazons" 
Argyrodes jxicunda id. ibid., p. 326, pi. 38, fig. 6. " 9 Amazons" 
Conopistha elongata Bryant, 1940, p. 306. figs. 68, 69, 75, 76. 

Male. Length, 2.5 mm., ceph. 1.2 mm., abd. 1.3 mm. 
Cephalothorax yellowish-brown, flat, quite long and about half as 
wide as long, thoracic groove transverse and deeply impressed, one- 



340 bulletin: museum of comparative zoology 

third nearer posterior margin than anterior, horn starts at margin of 
clypeus, very wide, narrowed gradually, tip not swollen, turned 
towards eyes, spoon-shaped, so that eye lobe apparently fits in it, 
about as wide as eye lobe and covered with short hairs, very little 
space between eye lobe and horn, ventral surface of horn convex; 
eyes, median eyes carried forward on a lobe, a.m.e. separated by a 
diameter, lateral eyes touching and at least three diameters of a.l.e. 
plus p.l.e. from space between horn and eye lobe; labium fused to 
sternum, wider than long; maxillae more than twice as long as labium, 
sides parallel; sternum yellowish-brown, triangular, widest between I 
pair of coxae, carried between II and III coxae and pointed in front of 
IV coxae, IV coxae touching; abdomen silvery, posterior tubercle 
rounded and directed backward, so that it extends beyond the plane 
of the spinnerets, a basal chitinized ring, easily seen in dorsal view, for 
stridulatory organ, carried on to ventral plate, venter infuscate with 
chitinized spots and lacking the usual silvery spots; palpus pale, 
terminal joint brown, as long as cephalothorax, patella swollen dor- 
sally, twice as long as tibia, cymbium with tip widened and corners 
produced as lobes, prolateral lobe with tip strongly chitinized, bifid, 
with margins rolled inward. 

Female. Length, 3.1 mm., ceph. 1.1 mm., abd. 2.1 mm. long, 
3.4 mm. high. 

Abdomen very high, not extending beyond the spinnerets as in the 
male, but rather upward, tip of abdomen contracted, venter very 
dark, with the usual pair of silver spots very faint, dark area extends 
on sides. O. P. Cambridge was undecided if lugens was the female of 
concinna, but he notes that both were from the same locality. The 
same is true of jucunda. 



Genus Theridion Walckenaer 1805 

The genus Theridion is based on a European species, and today it 
harbors a great many species that eventually will be placed in other 
genera. This particularly, is true of the species found in the West 
Indies. But until the fauna of more of the islands is known, it is 
thought best to leave these species in the genus as defined by Simon in 
the Histoire Naturelle des Araignees. 



BRYANT: SPIDERS OF THE VIRGIN ISLANDS 341 

Theridion brevipalpus spec. nov. 
Figures 28, 31, 32 

Male. Length, 1.2 mm. 

Ccphalofhorax yellow, shading to brown about the eyes, moderately 
high, smooth and shining, anterior margin not much narrowed, two 
very long bristles posterior to lateral eyes and a median row of three 
very long bristles, no thoracic groove; eyes cover the entire width of 
the head, anterior row slightly recurved, a.m.e. diurnal, slightly larger 
than a. I.e., separated by more than a radius and slightly nearer the 
a. I.e., lateral eyes touching, posterior row slightly procurved, each eye 
surrounded by a copper-red ring, p.m.e. largest of the eight, separated 
by a radius and from p. I.e. by a little more; quadrangle of median 
eyes higher than wide and narrower in front; clypeus about as high as 
quadrangle and convex; mandibles vertical, cone-shaped, fang long, 
fang groove poorly defined; labium brown, fused to sternum, wider 
than long, tip only slightly narrower than base and not rebordered; 
maxillae twice as long as labium, strongly inclined over labium; 
sternum pale, heart-shaped, strongly convex, IV coxae separated by 
more than a diameter; abdomen triangular, as wide as long, widest at 
basal third, tip pointed, dorsum flat, brown with no distinct pattern 
but blotches of cream-white and a darker brown about margins and 
three pairs of transverse dark bars on distal third, venter a pale brown, 
spinnerets inconspicuous; legs, IV left missing, anterior pairs very 
long, slender, all legs pale, femora with a wide dark band about middle 
that fades at distal and basal ends, distal joints almost white, with a 
broken ventral dark stripe on anterior tibiae, anterior femora slightly 
incrassate with two parallel ventral rows of small cusps, similar to 
Crushdina and at distal end, two widely separated rows of long slender 
bristles or spines, these bristles are from a distinct base and when 
broken, the scar can be seen, a long dorsal bristle at tip of patellae, a 
long dorsal bristle at middle of tibiae, no long bristles or spines on 
posterior pairs but rows of colorless hairs, tarsi about half as long as 
metatarsi, IV tarsus with a comb of 7 or 8 bristles; palpus very short, 
but little longer than the mandibles, femur, patella and tibia white, 
terminal joint brown, patella and tibia subequal, palpal organ fills the 
cavity, has no structure that can be seen, but has two slender dark 
spines at tip. 

Holotype cf St. Croix, (Beatty) 



342 bulletin: museum of comparative zoology 

The generic position of Theridion brcvipalpus is uncertain. It belongs 
to the genus Theridion in the sense used by Simon in "Histoire Natu- 
relle des Araignees", but it is undoubtedly a new genus as the legs 
have distinct spines and the palpus is very simple. 



Theridion cybele spec. nov. 
Figure 39 

Female. Length, 1.5 mm., ceph. 0.6 mm., abd. 0.9 mm. Ion , 
1.0 mm. wide. 

Cephalothorax pale, with a median gray stripe which starts posterior 
to the p.m.e. and does not reach the posterior margin, wider than long, 
no thoracic groove, rather high, posterior fifth falls rapidly to margin; 
eyes cover width of the head, anterior row recurved, a.m.e. diurnal, 
separated by more than a diameter, a.l.e. smaller and separated from 
a.m.e. by only a line, posterior row slightly procurved, same length as 
anterior, each eye surrounded by a red ring, equidistant, p.m.e. 
separated by fully a diameter and a little larger than p.l.e. ; quadrangle 
of median eyes, wider in front and higher than wide; clypeus convex, 
as high as quadrangle, margin dark; vmndihlcs; vertical, rather small, 
pale with a dark stripe on middle margin, no boss, fang weak, im- 
possible to see margin of fang groove; labium narrow, much wider 
than long, fused to sternum; maxillae more than twice as long as 
labium, very slightly incWned; sternum pale, convex, two-thirds as wide 
as long, truncate between IV coxae; abdomen wider than long, tri- 
angular, dorsum flat, a pair of divergent basal dark spots which soon 
fade into the white of the dorsum, many small dark dots, from each of 
which comes a long slender hair, near tip, a pair of converging dark 
stripes, venter infuscate, with curving dark stripes at base; legs, 
1-4-2-3, short, pale with dark ventral spots on all joints, distal spots 
heaviest, a slender dark spine at tip of patellae and middle of tibiae, 
IV tarsus with a comb of 6 or 7 curved bristles; epigynum rather large 
for so small a spider and protruding from the plane of the venter. 

Holotype 9 St. Croix, (Beatty) 

Paratype 9 St. Croix, (Beatty) 

This species does not belong to the genus Theridion in the strict 
sense, as the legs have spines and the epigynum is much more com- 
plicated than is usually found in that genus. Because of the spotted 
legs and the very broad abdomen with the small dark dots, it is a 
striking species. 



BRYANT: SPIDERS OF THE VIRGIN ISLANDS 343 

Theridion guanicae Petr. 
Figure 36 

Theridion guanicae Petrunkevitch, 1930. p. 208, figs. 57, 58. " 9 Porto Rico" 

This species was described from two females collected from under 
a rock in a cactus field near the coast at Guanica, Puerto Rico, January 
7, 1926 by Dr. Petrunkevitch. A male and a female were found by 
Beatty at St. Croix. These specimens agree very well with the descrip- 
tion except in a few minor points. The eyes are carried forward, best 
seen in a lateral view, and below the anterior row of eyes there is a 
groove shown in figure no. 57 as a straight line. The clypeus is convex; 
sternum convex, with the lateral margins slightly carried between the 
coxae, posteriorly the sternum is very wide, squarely truncate and 
carried between the coxae almost to the pedicel. The St. Croix female 
has the abdomen rather shrunken, so no markings can be traced. 

Male. Length, 1.0 mm. 

Cephalothorax same as in female, except for a pair of very long 
bristles below the a.m.e. directed anteriorly and four shorter bristles 
in a procurved row at the beginning of the thoracic slope; mandibles, 
tooth on the median margin of the fang groove larger than in female; 
sternum pale brown, truncate posteriorly, IV coxae separated by 
almost two diameters, lateral margins not carried between coxae; 
abdomen whitish, with a faint median gray stripe, scantily covered 
with long hairs, globose, venter pale; legs little paler than cephalo- 
thorax, rather short, little difference in length, I pair, tibia and meta- 
tarus with a prolateral and retrolateral rows of long, colorless bristles ; 
palpus short, in a ventral view near the tip on the prolateral side is a 
broad curved truncate piece that supports the embolus. 

Allotype cf St. Croix, (Beatty) 

9 St. Croix, (Beatty) 

The palpus indicates that this species belongs in the same section of 
the genus as Theridion unimaculatum Emerton. 

Theridion minutum (Petr.) 

Spintharus minutum Petrunkevitch, 1926, p. 51, fig. 11. " 9 St. Thomas"' 
Theridion dexteri Petr., 1930, p. 200, fi.gs. 45-50. 'V 9 Porto Rico, Campus of 
University at Rio Piedras" 

Male. Length, 1.7 mm. 
Female. Length, 2.2 mm. 

In both male and female, the head is much higher than is usual in 
the genus, thoracic groove near the posterior margin, usually marked 



344 bulletin: museum of comparative zoology 

by a black dash or line; a.m.e. diurnal, other eyes colorless; abdomen 
globose, base high, a dirty white, often marked with five small black 
dots, four in a quadrangle, followed by a fifth in the median line ; legs, 
anterior pairs very long, a black dot each side on patellae and tibiae, 
with many long colorless hairs and bristles and on the first tibia of 
both male and female at distal end, a prolateral row of four long 
spines and a ventral p. row of long bristles, all longer in the male 
than in the female; epigynum in all specimens very lightly chitinized, 
one specimen, (no. 22, Wilson coll.), shows distinctly the two pairs of 
circles shown by Petrunkevitch, 1926, fig. 11, but the posterior pair is 
small and apparently the openings, anterior pair larger and beneath 
the skin; palpus large for the size of the spider, agrees with the 
Petrunkevitch, 1930, figs. 48, 49, except that no tibial spur can be 
seen; this spur is figured as large and almost at right angles to tibia, 
however, the palpal organ agrees with the figure in all parts. 

d" 9 St. Croix, 28 March 1922, (Wilson) 

& 9 St. Croix, (Beatty) 

This species is not a Spintharus, since the posterior row of eyes is 
straight, and the genus is based on the strongly procurved posterior 
eye row. Neither is it a true Theridion as the palpus is totally unlike 
any in the genus and there are true spines on the legs of both male and 
female. 

In the collection sent by Dr. Wilson, there is a male and several 
females and one cocoon of eggs. The mesh of the cocoon is very loose, 
so that the eggs can be seen plainly. A note on the original label states 
that the specimens are always green, even the eggs, "always found as 
a single individual on a leaf. The web is very small, thin and scarcely 
visible. The spiders are easily captured. Species rather common." 

The species was described from a single female taken by Dr. Shoe- 
maker, July 1915 on St. Thomas. Through the kindness of Dr. Shoe- 
maker, I have been able to examine the type. The specimens collected 
by Wilson and Beatty on St. Croix, agree with it perfectly, and also 
with the description of Theridion dexteri Petr. 1930 from Puerto Rico. 

Theridion virginus spec. nov. 
Figure 27 

Male. Length, 1.8 mm. 

Cephalothorax pale brown, with a converging vague dark stripe from 
the lateral eyes to about the middle, lateral margins darker, four-fifths 
as wide as long, thoracic groove a slight transverse depression; eyes 



BRYANT: SPIDERS OF THE VIRGIN IvSLANDS 345 

cover width of head, each eye surrounded by a bright red ring, anterior 
row recurved, a.m.e. diurnal, separated by fully a diameter and from 
a. I.e. by a radius, lateral eyes touching, subequal, posterior row 
straight, eyes equidistant, p.m.e. separated by a scant diameter, p.l.e. 
smaller than p.m.e.; quadrangle of median eyes wider in front and as 
high as wide; clypeiis as high as quadrangle, a slight depression below 
anterior eye row and then convex to margin ; mandibles pale, long and 
attenuate, a slight swelling near base on lateral margin, fang groove 
short; labinm fused to sternum, triangular; maxillae one and a half 
times as long as labium, narrow, strongly inclined; sternum triangular, 
as wide as long, pale, margins dark; abdomen oval, twice as long as 
wide, dorsum flat, olive-brown with a pair of narrowly separated, 
interrupted white stripes, each side much darker from dark blotches, 
venter infuscate, epigastric fold extends beyond the middle, spin- 
nerets inconspicuous; legs, III left missing, 1-2-4-3, I pair very long, 
pale, with a narrow broken dark ring at tip of femur, a small dark spot 
on patella and at tip of tibia, no spines but rows of colorless hairs, 
other pairs pale with the same dark spots as on I pair; palpus rather 
short, pale, terminal joint darker, palpal organ filling cavity, with the 
same parts as in Theridion frondeum Hentz. 

Holotype cf St. Croix, (Beatty) 

Theridion virginus belongs near Theridion frondeum Hentz, which 
has been placed in the genus Phyllonethis. This genus has been 
synonymized with Theridion, but sometime it may be used as a 
sub-genus. Theridion frondeum was reported by Simon from St. Vin- 
cent, 1897. It is not impossible that his spider was the one here 
described, because frondeum is a northern species and has never been 
reported from any other island of the West Indies. 



Family ARGIOPIDAE 

Genus Argiope Audouin 1825 

Argiope argentata (Fabr.) 

Aranea argentata Fabricius, 1775 2, p. 414. " 9 India"; Petrunkevitch, 1926, 
p. 29. 

9 St. Croix, (Wilson, Beatty) 

Argiope trifasciata (Forskal) 

Aranea trifasciata Forskal, 1775, p. 40. " 9 Kahirae" 
cf 9 St. Croix, (Beatty) 



346 bulletin: museum of comparative zoology 

Genus Acacesia Simon 1895 

ACACESIA FOLIFERA (Marx) 
Epeira folifera Marx, 1890, p. 545. 

Epeira foliata Hentz, 1847, p. 475. " 9 Alabama"; preoccupied by Walcken- 
aer, 1837, 2, p. 62. 

2 9 St. Croix, (Beatty) 

Genus Cyclosa Menge 1866 

Cyclosa oculata (Walck.) 

^ranea ocwZa/o Walckenaer, 1802, p. 428. "cf 9 Paris" 
Cyclosa walckenaeri Petrunkevich, 1926, p. 29. 
o" 9 St. Croix, (Beatty) 

Genus Eustala Simon 1895 

EUSTALA ANASTERA (Walck.) 

Epeira anastera Walckenaer, 1837, 2, p. 33. "Georgia, Abbot, Georgia spiders, 
fig. 381" 

cf 9 St. Croix, (Beatty) 

Genus Eriophora Simon 1894 

Eriophora edax (Blackwall) 

Epeira edax Blackwall, 1863, p. 30. 'V 9 Rio Janeiro" 
9 St. Croix, (Beatty) 

Genus Metepeira F. O. P. Cambridge 1903 
Metepeira labyrinthea (Hentz) 

Epeira labyrinthea Hentz, 1847, p. 471, pi. 31, fig. 3. "9 North Carolina, 

Alabama" 
Araneus {Metepeira) labyrintheus Petrunkevitch. 1926, p. 27, "St. Croix, 

common on telephone wires" 

cT 9 St. Croix, (Wilson, Beatty) 

Genus Neoscona Simon 1864 
Neoscona oaxacensis (Keys.) 

Epeira oaxacensis Keyserling, 1863, p. 121. " 9 Oaxaca, Mexico" 
d" 9 St. Croix, (Beatty) 



BRYANT: SPIDERS OF THE VIRGIN ISLANDS 347 

Genus Nephila Leach 1815 

Nephila clavipes (Linn.) 

Aranea clavipes Linnaeus, 1767, p. 1034, no. 27. " 9 in America" 
Nephila clavipes Petrunkevitch, 1926, p. 30. , 

9 St. Croix, (Beatty) 

Genus Alcimosphenus Simon 1895 

Alcimosphenus licinus Simon 

Alciinosphonis licinus Simon, 1895, p. 951. " 9 Jamaica et S. Domingo" 
9 St. Croix, (Beatty) 

Genus Leucauge White 1841 

Leucauge argyra (Walck.) 

Tetragnatha argyra Walckenaer, 1837, 2, p. 219. "cf 9 La Guadeloupe" 
Leucauge argyra Petrunkevitch, 1926, p. 29. "common, St. Thomas" 
d^ 9 St. Croix, (Beatty) 

Leucauge regnyi (Simon) 

Argyroepeira regnyi Simon, 1897, p. 871. "c? 9 St. Vincent" 
Leucauge regnyi Petrunkevitch, 1926, p. 30. " 9 St. Thomas" 
immat. 9 St. Croix, (Beatty) 

Genus Tetragnatha Latreille 1804 

Tetragnatha antillana Simon 

Tetragnatha antillana Simon, 1897, p. 868. "cf 9 St. Vincent" 
& 9 St. Croix, (Beatty) 

Tetragnatha piscatoria Simon 

Tetragnatha piscatoria Simon, 1897, p. 869. "cf St. Vincent" 
cf St. Croix, (Beatty) 

Genus Wendilgarda Keyserling 1886 

Wendilgarda theridionina Simon 

Vendilgarda theridionina Simon, 1895, p. 919, fig. 986. "cf 9 Venezuela; 
San-Esteb^n" 

9 s St. Croix, (Beatty) 



348 bulletin: museum of comparative zoology 

Genus Gasteracantha Sundevall 1833 

Gasteracantha tetracantha (Linn.) 

Aranea tetracantha Linnaeus, 1767, p. 1037, no. 45. " 9 St. Thomas" 
Gasteracantha tetracantha Petrunkevitch, 1926, p. 29. "cf 9 St. Thomas" 
9 s St. Croix, (Beatty) 



Family GNAPHOSIDAE 

Genus Teminius Keyserling 1887 

Teminius insularis Keys. 

Teminius insularis Keyserling, 1887, p. 422, pi. 6, fig. 1. "9 Hayti" 
Syrisca hirsuta Petrunkevitch, 1926, p. 63, figs. 20, 21, cf Sta. Maria Bay, 

St. Thomas, July 28; nee. Petrunkevitch, 1925, p. 151, figs. 74-76. 
Eutychuroides fusca Petrunkevitch, 1926. p. 57, fig. 17. "2 9 St. Thomas" 

Through the kindness of Dr. Shoemaker, I have been able to 
examine one of the females of Eutychuroides fusca that was used by 
Dr. Petrunkevitch in his description. It is probably one moult from 
maturity and was killed soon after moulting, as the spinnerets are 
extended and apparently longer than in an adult specimen. The 
maxillae are very faintly impressed, but the specimen agrees in all 
other respects with the Keyserling type from Haiti. 

Dr. Petrunkevitch gives a long, detailed description of the male 
found on St. Thomas and calls attention to the difference in the eyes 
and the proportion of the various parts of the palpus from the male 
described by him from Panama. He also calls attention to the long 
posterior pair of spinnerets, and the faint depression on the maxillae. 

The male and female differ greatly in size and the size of the spines 
on the anterior pairs of legs. In the female, the three pairs of ventral 
spines on the first tibia are so slender that they are easily overlooked 
or recognized only as bristles, in the mass of hairs that cover the joint; 
this is particularly true of the apical pair. In the male, the covering 
of hairs is not as thick, and the spines are heavy and much longer than 
the diameter of the joint, so they are very conspicuous. The male and 
female have the same pale stripe on the cephalothorax and the basal 
half of the abdomen, with chevrons on the posterior half. 

9 St. Croix, 30 March 1922, (Wilson) 

c^ 9 St. Croix, (Beatty) 



BRYANT: SPIDERS OF THE VIRGIN ISLANDS 349 

Family SPARASSIDAE 

Genus Heteropoda Latreille 1904 

Heteropoda venatoria (Linn.) 

Araiua venatoria Linnaeus, 1767, p. 1035. "in America calidiore" 
Heteropoda venatoria Petrunkevitch, 1926, p. 30. 
cf s 9 s St. Croix, (Beatty) 

Genus Olios Walckenaer 1837 

Olios antiguensis (Keys.) 

Sparassus antiguensis Keyserling, 1880, p. 264, pi. 7, fig. 146. "cf 9 Antigua" 
cf 9 St. Croix, (Beatty) 

Genus Pseudosparianthis Simon 1887 

Pseudosparianthis antiguensis Bryant 

Pseudosparianthis antiguensis Bryant, 1923, p. 13, pi. 1, fig. 4. " 9 Antigua" 
2 9 St. Croix, (Beatty) 



Family SELENOPIDAE 

Genus Selenops Latreille 1819 

Selenops lindborgi Petr. 

Figures 29, 37 

Selenops lindborgi Petrunkevitch, 1926, p. 55, fig. 16. " 9 Sta. Maria Bay, 
St. Thomas, July 28" 

Male. Length, 6.5 mm., ceph. 3.3 mm. long, 3.6 mm. wide, abd. 
3.0 mm. long, 2.6 mm. wide. 

Cephalothorax pale yellow, darker about the margin, with a row of 
slender hairs or bristles on margin directed forward, wider than long, 
very flat, thoracic groove longitudinal and faint, radial striae distinct 
near groove but disappearing before margin; eyes in three groups, four 
central and two in each lateral group, all eyes heavily ringed in black, 
eyes of median group in a recurved row, central eyes separated by 
about three-quarters the diameter of the next eye, posterior lateral 
eye on a large rounded tubercle, largest of the eight and directed back- 
ward, inferior lateral eye very small, about one-half diameter of 
central eye and about midway between median eye and p. I.e. ; clypeus 



350 bulletin: museum of comparative zoology 

less than a radius of central eye, with a scant fringe of long hairs on 
margin; mandibles vertical, rather small, yellow with stripes of dull 
gray, boss small and inconspicuous, fang groove oblique, superior 
margin with three widely separated teeth, inferior margin with two 
teeth ; labium yellow, longer than wide, tip rounded, slightly excavate 
on lateral margins at base; maxillae parallel, fully twice as long as 
labium; sternum pale, slightly wider than long, bifid between IV 
coxae; abdomen pale, with lateral and posterior margins gray, flat, 
anterior margin truncate, venter pale; legs, I, II and IV left and III 
right missing, yellow with no indications of the wide dark bands 
found in the female, spines much shorter than in female, spines, I 
pair, patella, 0, tibia, dorsal, 1, prolateral, 2, retrolateral, 2, ventral, 
2-2-2-2, distal pair very small, metatarsus, dorsal, 0, prolateral, 1, 
retrolateral, 0, ventral, 2-2, II pair, tibia, dorsal, 2, prolateral, 2, 
retrolateral, 2, ventral, 2-2-2, metatarsus, prolateral, 2, retrolateral, 
1, ventral, 2-2; palpus shorter than cephalothorax, pale, tibia little 
longer than patella, tibia with no dorsal apophysis, large lateral 
apophysis as figured. 

Allotype cf St. Croix, (Beatty) 

3 9 St. Croix, (Beatty) 

It is not impossible that the first leg of the male has been renewed 
recently, as it is shorter than the second leg and the spines are much 
smaller. 

It is with some hesitation that these specimens are identified as 
Selenops lindborgi Petrunkevitch. The type specimen is from St. 
Thomas and it is described as 12.3 mm. long, but it may not be adult. 
In some genera, the penultimate stage is longer than the adult. The 
three females from St. Croix are all smaller than the dimensions given 
for the type, as the largest is only 10.0 mm. long, but they agree with 
the description in the size and position of the eyes and in the spines. 
The epigynum of one of the specimens is figured. This differs some- 
what from the figure of the type, but F. O. P. Cambridge in the Biol. 
Centr. America, 1900, 2, p. 117, says, "The vulva in Selenops varies 
very much in detail, so that many specimens obviously the young of 
S. mexicanus appear at first sight to belong to entirely different 
species." 

Selenops longipes Petrunkevitch from Puerto Rico, known only from 
the male, is very near S. lindborgi. Dr. Gertsch has kindly sent me 
drawings of the eyes and the palpus of the type specimen in the 
American Museum. The median eyes are round, not ellipical as 
figured and the median eyes touch the neighboring eye as stated in 



BRYANT: SPIDERS OF THE VIRGIN ISLANDS 351 

the text. In the palpus, the long process at the base of the embolus is 
folded longitudinally and one corner of the tip is prolonged in a sharp 
point. The various parts of the palpus are very similar in the two 
species. The greatest diflerence is in the eyes. It is not impossible 
that the smaller form of S. lindborgi reported by Dr. Petrunkevitch 
from Puerto Rico is the female of S. longipes. 



Family THOMISIDAE 

Genus Misumenops F. O. P. Cambridge 1900 

MisuMENOPS BELLULUS (Banks) 

Misumena bellula Banks, 1896, p. 71. "9 Funta Gorda, Fla." 
12 cT St. Croix, (Beatty) 

Family CLUBIONIDAE 

Genus Chiracanthium C. Koch 1839 

Chiracanthium inclusum (Hentz) 

Clitbiona inclusa Hentz, 1847, p. 51, pi. 23, fig. 18. " 9 South Carolina, North 
Carolina, etc." 

c? 2 9 St. Croix, (Beatty) 

Genus Aysha Keyserling 1891 

Aysha tenuis (L. Koch) 

Anyphaena tenuis L. Koch, 1866, p. 211, pi. 9, fig. 140. " 9 St. Domingo" 
cf 9 St. Croix, (Wilson) 
9 cf 8 9 St. Croix, (Beatty) 

Genus Corinna C. Koch 1842 

Corinna antillana spec, no v. 

Figure 33 

Female. Length, 11.2 mm. without mand., ceph. 5.4 mm. long, 
3.5 mm. wade, abd. 5.4 mm. 

Ccphahthorax mahogany-red, darker about eyes, roughened, 
thoracic groove very distinct; eyes cover three-quarters width of head, 
anterior row straight, a.m.e. largest of the eight, separated by two- 
thirds of a diameter and from a.l.e. by a diameter and a half, posterior 



352 bulletin: museum of comparative zoology 

row straight, little longer than anterior row, eyes subequal, p.m.e. 
separated by a diameter and a half and from p.l.e. by two diameters 
and a half; quadrangle of median eyes slightly wider in front and not 
as high as wide; dypeus less than a radius of a.m.e.; mandibles darker 
than cephalothorax, roughened, geniculate, boss large, fang groove 
short, superior margin with three teeth, inferior margin with three 
large teeth followed by one much smaller, fang short with a thick base; 
lahinvi about as wide as long; maxillae not quite twice as long as 
labium, lateral margins almost parallel ; sternum rebordered on lateral 
margins, two-thirds as wide as long, IV coxae separated by half a 
diameter; abdomen a gray-brown, slightly darker at base, cylindrical; 
legs, 4-1-2-3, yellow with distal joints darker, spines, I pair, patella, 
0, tibia, dorsal, 0, lateral, 0, ventral, 2-2-2-2-2, metatarsus, dorsal, 0, 
lateral, 0, ventral, 2-2, with a single median spine at tip, II pair, 
tibia, ventral, 2-2-2, Ir, metatarsus, ventral, 2-2, 1 median apical 
spine, all metatarsi with ventral apical spine; epigynum, large, pos- 
terior portion strongly chitinized, anterior portion convex. 

Holotype 9 St. Croix, (Beatty) 

Two other species of Corinna have been reported from the Virgin 
Islands. C. humilis (Keys.), known from both male and female, has 
three pairs of ventral spines on the first tibia and two pairs on the 
metatarsus. C. cleonei Petr. known only from the type( 9 ), has four 
pairs of spines on the first tibia and two pairs on the metatarsus. C. 
antillana has but three large teeth and one small tooth on the inferior 
margin of the fang groove but it has five pairs of spines beneath the 
first tibia and two pairs on the metatarsus and a small median apical 
spine. Simon described seven species of Corinna from St. Vincent, and 
one has the single apical spine on the metatarsi, although several have 
three pairs of spines on the first metatarsus. 



Corinna humilis (Keys.) 

Hypsinotus hum.ilis Keyserling, 1887. p. 446, pi. 6, fig. 18. 'V 9 St. Kitts" 
Corinna humilis Petrunkevitch, 1926, p. 30, 4 9 
2 9 St. Croix, (Beatty) 



BRYANT: SPIDERS OF THE VIRGIN ISLANDS 353 

Family SALTICIDAE 

Genus Sidusa Peckham 1895 

SiDUSA pavida spec. nov. 

Figures 34, 40 

INIale. Length, 3.0 mm., ceph. 1.6 mm., abd. 1.5 mm. 

Cephalothorax brown, almost black about the eyes, with a paler 
median stripe from dorsal eyes to posterior margin, eye area thickly 
covered with short white hairs, thoracic groove very short, in a small 
depression slightly posterior to dorsal eyes, cephalothorax moderately 
high, highest at dorsal eyes where it continues in same plane half way 
to margin and then falls rapidly, sides vertical; eyes, anterior row re- 
curved, a.m.e. twice the diameter of a.l.e., separated by a line and 
from a.l.e. by a little more, second row midway between first and 
third rows, dorsal eyes and a.l.e. subequal, not quite on extreme 
margin of carapace; quadrangle of eyes about two-fifths as wide as 
long and narrower behind; clypeus inclined inward, pale, covered with 
white scales, almost as wide as diameter of a.m.e.; mandibles pale, 
vertical, anterior surface flat, fang groove very short, transverse, 
superior margin with two small, black subequal teeth at median margin, 
best seen from front view, inferior margin not defined and with no 
tooth, fang from a thick base and longer than groove; labium pale, 
longer than wide; maxillae twice as long as labium; sternum small, 
truncate at both ends, two-thirds as wide as long, convex, pale shaded 
with gray; abdomen pale, with a pair of dark lateral stripes with an 
irregular median margin which forms three pairs of white spots, the 
largest of which is posterior to middle, median area and sides covered 
with white hairs, venter pale; legs, 4-1-2-3, all patellae with pro- 
lateral and retrolateral spines, anterior tarsi and metatarsi subequal, 
pale, I pair with a broad prolateral dark stripe on patella and tibia, 
spines, I pair, tibia, dorsal, 0, prolateral, 2, retrolateral, 1 small basal, 
ventral, 2 apical, 2 median, Ir, metatarsus, dorsal, 0, prolateral, 0, 
retrolateral, 0, ventral, 2-2, basal pair much more than half the length 
of the joint, II pair, tibia, dorsal, 0, prolateral, 2, retrolateral, 0, 
ventral, 2, Ir, Ir, metatarsus, ventral, 2-2, III and IV pairs spiny, 
tibiae with dorsal basal spine; palpus, basal joints pale, not as long 
as cephalothorax, femur not enlarged, tibia longer than patella, 
swollen so that it is almost as wide as long, prolateral half covered with 
short, black hairs, retrolateral with long white hairs, tibial apophysis 
long, parallel to cymbium, suddenly contracting and ending in a long 
bristle like tip, embolus a sHghtly curved spine on opposite side from 
tibial apophysis. 



354 bulletin: museum of comparative zoology 

Holotype cf St. Croix, (Beatty) 

The generic position of pavida is uncertain, but it can be placed in 
the genus Sidusa as construed by F. O. P. Cambridge in the Biol. 
Centr. America, 1901, 2, p. 196. In it, he placed several of the 
Peckham genera, irrespective of the number of teeth on the inferior 
margin of the fang groove, but always with two teeth on the superior 
margin. According to Simon, all tibiae in the genus Sidusa have a 
dorsal basal spine. Cambridge states that the spine is always found 
on the posterior pairs but is often missing on the anterior. This 
species cannot be placed in the genus Stoidis Simon, 1901, as the III 
and IV pairs of legs are not subequal and the clypeus is fairly high 
and covered with hairs. 



Genus Plexippus C. Koch 1850 

Plexippus paykulli (Audouin) 

Attus paykulli Audouin, in Savigny, Descr. Egypte, Nat. Hist., 1826, 1, p. 172. 

"c? Egypte" 
Plexippus paykulli Petrunkevitch, 1926, p. 21, 31. 
a" St. Croix, 14 March 1922, (Wilson) 



Genus Corythalia C. Koch 1850 

CORYTHALIA IRIDESCENS Pctr. 

Cor;/</iaKa mdescens Petrunkevitch, 1926, p. 69, figs. 23, 24. 'V 9 St. 
Thomas" 

Through the kindness of Dr. Shoemaker, I have been able to examine 
the types. The species belongs to the genus Corythalia in the broad 
sense, as the third pair of legs are longest and they have a heavy 
fringe of hairs, but the embolus is not curved. It is apparently subject 
to great variation in size. The patellae are flattened dorsally, possibly 
more in the female than in the male. In the male palpus, the patella 
is half as long as the femur, dorsally flattened and thickly covered with 
white iridescent scales, tibia seen from above, is less than half as 
long as patella, with a slender retrolateral apophysis and a ventral 
apophysis that is not quite at the tip, half hidden by dark hairs and 
best seen from a lateral view. 



BRYANT: SPIDERS OF THE VIRGIN ISLANDS 355 

Genus Hentzia Marx 1883 

Hentzia antillana Bryant 

Hentzia antillana Bryant, 1940, p. 494, figs. 285, 289, 294. "d" 9 Antigua" 
d" 9 St. Croix, (Beatty) 

Genus Stridulattus Petrunkevitch 1926 

Stridulattus stridulans Petr. 

Stridulattus stridulans Petrunkevitch, 1926, p. 74, figs. 25-28. 'V St. Thomas, 
July 1915" 

Through the kindness of Dr. Shoemaker, I have been able to 
examine the type, the only specimen known. 

The genus has been misplaced and it belongs near Marpissa, rather 
than Hcliophanes. The diagnosis is misleading; the first coxae are 
described as far apart but in the description of the species, the coxae 
are separated by the width of the labium. The latter is correct, as 
they are separated by a little over a diameter. The sternum is dis- 
tinctly narrowed between the first coxae. In the Biol. Centr. America, 
the relative length of the sternum is used to separate genera, but this is 
a character that does not always hold in all species. There is also a 
reduction of spines on all legs, both in size and number. On the right 
first leg, the tibia has one retrolateral spine, a ventral apical pair 
followed by two on the inner side; the left leg has an additional spine 
opposite the inner basal spine. All the spines are small and the 
posterior metatarsi have an apical whorl only. 

Genus Menemerus Simon 1868 

Menemerus kochi spec. nov. 

Figure 30 

Female. Length, 4.5 mm., ceph. 2.4 mm., abd. 2.1 mm. 

Cephalothorax dark brown, much darker about the eyes, with a 
rather broad lateral stripe that does not meet at posterior margin, a 
median stripe of white hairs from a little anterior to thoracic groove to 
near the posterior margin, moderately high, widest posterior to dorsal 
eyes, cephalic portion rather flat, a shallow depression between dorsal 
ej^es in which is the short thoracic groove; eyes, anterior row slightly 
recurved, eyes equidistant, a.m.e. large, separated by little more than 
a line, a.l.e. less than a radius of a.m.e., second row nearer the first 
than the third row, dorsal eyes convex, subequal with a.l.e., not on 
extreme margin of carapace; quadrangle as wide in front as behind; 



356 bulletin: museum of comparative zoology 

clypeus thickly covered with white hairs, narrow, less than a radius 
of a.m.e. ; mandibles brown, robust, vertical, fang groove slightly 
oblique, superior margin with two contiguous black teeth, inferior 
margin with one black tooth, fang little longer than groove with a 
thick base; labium longer than wide; maxillae almost twice as long as 
labium; sternum anteriorly narrowed to width of labium, which is less 
than a diameter of I coxa, widest between II and III coxae, where it is 
about half as wide as long, truncate in front of IV coxae; abdomen 
oval, dorsum flat, covered with short pale hairs and longer black ones, 
a wide white basal band which does not extend on sides, posterior part 
an orange-red, a white median stripe that just touches basal band, 
widest posterior to middle, with posterior third much narrowed and 
broken by three dark chevrons, sides dark, venter pale with dark spots 
in a median and lateral stripes, area between, with scattered dark spots, 
largest a pair just anterior to spinnerets, spinnerets long and closely 
grouped; legs, 4-1-2-3, I pair enlarged, femur flattened laterally, 
femur pale with a wide dark distal band, patella, tibia and metatarsus 
brown, tarsus pale, spines, patella, 0, tibia, dorsal, 0, lateral, 0, ventral, 
2-2-2, metatarsus, dorsal, 0, lateral, 0, ventral, 2-2, II pair colored as 

1 pair, spines, patella, prolateral, 1, tibia, ventral, 2-2-lr, metatarsus, 
ventral, 2-2, III and IV pairs, brown, spines, patella, prolateral, 1, 
retrolateral, 1, tibia, scattered, no dorsal, basal spine, metatarsus, 

2 whorls, apical and median; palpi pale, with a distinct dark spot 
at base of three terminal joints; epigynum a pair of transverse oval 
openings anterior to dark sacs beneath the skin, a median inverted 
chitinized V above the fold separates the sacs. 

Holotype 9 St. Croix, (Beatty) 

In 1846, C. Koch described Marpissa incerta ( 9 ) from St. Thomas. 
This has never been found since. It is a little larger than this species, 
and the description was made when the specimen was dry. It varies 
in several points other than size, and probably it is not Menemerus 
kochi. Koch describes and figures this species with narrow lateral 
stripes on the cephalothorax and a wider stripe on anterior portion 
ending below the dorsal eye, no median stripe. The abdomen is all 
dark with a narrow lateral white stripe from base to spinnerets and 
the legs reddish-yellow. No mention is made of the conspicuous dark 
spots on the palpi. 



BRYANT: SPIDERS OF THE VIRGIN ISLANDS 357 

Genus Habronattus F. O. P. Cambridge 1901 

Habronattus brunneus var. insignis var. nov. 

Figures 35, 38 

Male. Length, 4.6 mm., eeph. 2.6 mm., abd. 2.3 mm. 

Cephalothorax, ocular area thickly covered with white scales and 
longer dark hairs, sides chestnut-brown, a median brown triangle from 
posterior margin with apex at groove, each side are inverted triangles 
of white scales, a narrow marginal line of white scales from below 
dorsal eyes, sides rounded, widest posterior to dorsal eyes, a recurved 
depression posterior to eyes; eyes, anterior row straight by upper 
margins, a.m.e. almost touching, a. I.e. about a radius of a.m.e. and 
separated from them by twice the space between a.m.e., small eyes 
about midway between first and third rows, dorsal eyes not on ex- 
treme margin of carapace and subequal with a.l.e. ; clypcus covered 
with white scales, about a diameter of a.m.e. and retreating; mandibles, 
small, cone-shaped and weak, inferior margin with one sharp tooth; 
abdomen oval, dorsum flat, dark, with a broad basal white band that 
narrows on sides and reaches the spinnerets, a median basal spear 
mark of white scales that reaches middle, where it connects with a pair 
of narrow transverse bars, a median white triangle on posterior half 
and two pairs of widely separated lateral white dots, between basal 
mark and posterior triangle an indistinct pale chevron, venter pale 
with a median dark stripe that is widest anterior to spinnerets and 
heavier lateral stripes that do not reach the middle; legs, 3-4-1-2, 
with I and III pairs modified, all patellae with prolateral and retro- 
lateral spines, I pair dark, with a median stripe of white scales on 
femur and patella, femur with a prolateral brush of dark hairs with 
longer white pedicellate hairs, heaviest at distal half and a smaller 
brush on retrolateral side, ventral side with pedicellate white hairs, 
patella with prolateral and retrolateral tufts of hairs, tibia with scat- 
tered dorsal white scales, a prolateral and a retrolateral brush of dark 
hairs with longer white pedicellate hairs, metatarsus and tarsus dark, 
spines, tibia, with two long, heavy, prolateral fusiform spines, meta- 
tarsus, ventral, 2-2, II pair, dark with white scales. III pair, femur 
pale, with a large prolateral dark spot from base to near tip, tip 
swollen, pale, with a prolateral dark spot that bares a spine, swollen 
area with a thick dorsal crest of fawn-colored hairs, patella pale, 
dorsal, distal end modified by two processes, as in coronatus, distal 
process very slender, spines, tibia, dorsal, small median basal spine, 
ventral, a long, median basal colorless spine; palpus, pale, with white 
scales, tibia shorter than patella, tibial apophysis dark, heavy with a 



358 bulletin: museum of comparative zoology 

blunt tip, palpal organ of usual type, bulb convex, circular with the 
two processes starting near the base on prolateral side, the outer 
ending at tip. 

Female. Length, 7.0 mm., ceph. 3.1 mm., abd. 4.0 mm. 
Cephalothorax brown, covered with white scales, thickest on ocular 
area, where there are also scattered long black hairs, no indications of 
the white triangles or marginal line found in the male, sides rounded, 
recurved depression posterior to dorsal eyes; eyes same as in male; 
clypeus thickly covered with white scales; mandibles same as in male; 
abdomen brown, thickly covered with white scales and longer dark 
hairs, the posterior white triangle and the posterior pair of lateral 
spots can be traced, venter, pale with the three dark stripes as in the 
male; legs, 3-4-1-2, not modified, I pair heaviest, brown with scattered 
white scales, spines, patella, 0, tibia, dorsal, 0, lateral, 0, ventral, 
2-2-2, not opposite, retrolateral spines very short, metatarsus, dorsal, 
0, lateral, 0, ventral, 2-2, other pairs paler, II patella, 1 retrolateral 
spine. III tibia, dorsal, no median basal spine, ventral, a median basal 
spine, dark and not as long as in the male, posterior patellae with 
prolateral and retrolateral spines; epigynum with a median transverse 
opening followed by diverging dark areas beneath the skin, as in 
coronatus. 

Holotype d^ St. Croix, summer of 1941, (Beatty) 
Allotype 9 St. Croix, summer of 1941, (Beatty) 
Paratypes 3cf St. Croix, summer of 1941, (Beatty) 
The type of Habronattus hrunneus is from Key West, Florida and is 
in the museum collection. The variety insignis agrees in many ways 
with the type but differs from it principally in color and the secondary 
characters. The Florida specimen is brown, with no lateral triangles 
of white hairs on the cephalothorax, possibly from age, although there 
is no mention of them in the original description. The sternum, 
coxae and venter are dark brown, so that the four white stripes on the 
venter are quite conspicuous. It also has a much denser row of dark 
hairs above the anterior eyes. The variety insignis has an almost 
white sternum, coxae and venter and the four pale stripes mentioned 
by Peckham are poorly defined. But the difference is most notice- 
able in the secondary characters; the lateral brushes of dark hairs on 
the first tibia are heavier, the third leg is longer, with the distal end of 
the femur more swollen with a larger crest of fawn-colored hairs and 
the apophyses on the patella are larger. Both have the long colorless 
ventral spine on the third tibia and the bulb of the palpus is convex, 
rather than flat. The ventral spine on the third tibia was not noted in 
earlier descriptions. The female has never been described before. 



BRYANT: SPIDERS OF THE VIRGIN ISLANDS 359 

Unfortunately the original figure of the palpus of hrunneus is inaccu- 
i-ate. There are two dark spines, as is usually found in the genus, that 
closely follow the contour of the cavity and are parallel. The drawing 
shows but one. The tibial apophysis is dark and heavy. 

APPENDIX 

Other Spiders Reported from the Virgin Islands 

For the sake of completeness, all other spiders reported from the 
Virgin Islands are given below in a systematic list. In no case have the 
specimens been examined. An asterisk (*) before the name indicates 
one of the Virgin Islands as the type locality. In every case, the 
author and the year of the description are given. A second name and 
date is a reference to the person reporting the species from there. 

Sub-order MYGALOMORPHAE 
Family CTENIZIDAE 
Phaeoclita fauna Simon, 1889; Petrunkevitch, 1926 

Family DIPLURIDAE 

*Mygale drassiformis C. Koch, 1842 "St. Thomas" 
*DiPLURA MACRURA (C. Koch), 1842 "St. Thomas" 

Family BARYCHELIDAE 
*Obaerarius insulanus Petr., 1926 " 9 St. Thomas" 

Family THERAPHOSIDAE 

*Cyrtopholis acutispina Strand, 1907 "St. Thomas" 
Cyrtopholis bartholomei (Latreille), 1832; Petrunke\'itch, 1926 
*LscHNOcoLUS SHOEMAKERi Petr., 1926 "9 St. Thomas" 

Sub-order ARACHNOMORPHAE 
Family DYSDERIDAE 
*Ariadxa arthuri Petr., 1926 "immat. 9 St. Thomas" 

Family SICARIIDAE 
LoxoscELES RUFiPES (Lucas), 1834; Petrunkevitch, 1926 



360 bulletin: museum of comparative zoology 

Family PHOLCIDAE 
MoDisiMus glauca Simon, 1893; Petrunkevitch, 1926 

Family THERIDIIDAE 
Theridion tepidariorum C. Koch, 1841; Petrunkevitch, 1926 

Family ULOBORIDAE 
*Miagrammopes ciliatus Petr., 1926 " 9 St. Thomas" 

Family ARGIOPIDAE 
Araneus nauticus (L. Koch), 1875; Petrunkevitch, 1926 
Leucauge mandibulata F. O. p. Cambr., 1903; Petrunkevitch, 1926 
Micrathena militaris (Fabr.), 1775; Petrunkevitch, 1926 
Micrathena sagittata (Walck.), 1837; Petrunkevitch, 1926 

Family CTENIDAE 
*Odo agilis Simon, 1896 "St. Thomas" 

Family GNAPHOSIDAE 
*Trochanteria ranumcula Karsch, 1879 "cf 9 Sta. Cruz." 

Family SPARASSIDAE 
Olios fasciculatus Simon, 1880; Petrunkevitch, 1926 

Family CLUBIONIDAE 
*Clubiona maritima L. Koch, 1866 " 9 St. Thomas" 
*CoRiNNA cleonei Petr., 1926 "9 St. Thomas" 

Family SALTICIDAE 
*EvoPHRYS VETUSTA C. Koch, 1846 "immat. 9 St. Thomas" 
*Marpissa incerta C. Koch, 1846 "9 St. Thomas" 
Menemerus bivittatus (Dufour), 1831; Petrunkevitch, 1926 



BRYANT: SPIDERS OF THE VIRGIN ISLANDS 361 



BIBLIOGRAPHY 
AuDotriN, V. 

1827. Explicat. sommaire des pi. d'arachnides. In Savigny, Description 
de I'Egypte, Nat. Hist., 1, pt. 4. 

Banks, Nathan 

1896. New North American Spiders and Mites. Trans. Amer. Ent. 
Soc, 23, pp. 57-77. 

1901. Some Spiders and other Arachnids from Porto Rico. Proc. U. S. 
Nat. Mus., 24, pp. 217-227, pi. 15. 

Berland, L. et Fage, L. 

1914-1937. Les Arachnides de France, 6, pp. 1-1298. 

Blackwall, John 

1863. Descriptions of newly discovered Spiders captured in Rio Janeiro, 
by J. Gray and H. Clark. Ann. Mag. Nat. Hist., 11, pp. 29-45. 

Bryant, Elizabeth B. 

1923. Report on the Spiders Collected by the Barbados- Antigua Ex- 
pedition. Univ. Iowa Studies, Nat. Hist., 10, pp. 10-16, pi. 1. 

1940. Cuban Spiders in the Museum of Comparative Zoology. Bull. 
Mus. Comp. Zool., Harvard, 86, pp. 247-532, pis. 1-22. 

Cambridge, O. P. 

1880. On some new and little known Spiders of the Genus Argyrodes. 
Proc. Zool. Soc. London, pp. 320-342, pis. 28-30. 

Chamberlin, Ralph V. 

1917. New Spiders of the Family Aviculariidae. Bull. Mus. Comp. 
Zool., Harvard, 61, pp. 25-75, pis. 1-5. 

Fabricius, Joh. Christ. 

1775. Systema Entomologiae, etc. 1 volume. Aranea, pp. 431-429. 

FORSKAL, P. 

1775 insectorum, etc. Havniae, 1775. Descriptiones animalium. 

Hentz, N. M. 

1842. Descriptions and Figures of the Araneides of the United States. 

Joum. Boston Soc. Nat. Hist., 4, pp. 223-231, pi. 8. 
1845. (continuation). 5, pp. 189-202, pis. 16-17. 
1847. (continuation). 5, pp. 444-478, pis. 23-24. 
1850. (continuation). 6, pp. 271-295, pis. 9-10. 
1875. The Spiders of the United States. Occ. Papers Boston Soc. Nat. 

Hist., 2, pp. XIIH-171, pis. 1-21. 



362 bulletin: museum of comparative zoology 

Karsch, F. 

1879. Arachnologische Beitrage. Zeitschr. Gesamm. Naturwiss., 4, 
pp. 534-562, pi. 7. 

Keyserling, Graf Eugene 

1863. Beschreibung neuer und wenig bekannter Arten aus der Familie 
Orbitelae, etc. Sitz-Bericht. Isis, Dresden, pp. 63-154, pis. 1-7. 

1865. Beitrage sur Kenntniss der Orbitelae. Verh. k. k. Z. B. Gesell. 
Wien, 15, pp. 799-856, pis. 18-21. 

1887. Neue Spinnen aus Amerika. Ibid., 37, pp. 421-490, pi. 6. 

1880-1891. Die Spinnen Amerikas. In 4 volumes, published at Niirn- 
berg. Vol. 1, Laterigradae, 1880, pp. 1-283, pis. 1-8. Vol. 2, 
Theridiidae, in 2 parts, 1884 + 1886, pp. 1-222, pis. 1-10. and 
pp. 1-245, pis. 11-21. Vol. 3, Brasilianische Spinnen, 1891, pp. 
1-278, pis. 1-10. 

Koch, C. 

1836-1848. Die Arachniden, etc., vols. 3-16. 

Koch, Ludwig 

1886-1887. Die Arachniden-Familie der Drassiden. pp. 1-352, pis. 1-14. 

Linnaeus, C. 

1766-1768. Systema Naturae, etc. editio XII. 

Lucas, H. 

1845. Sur une nouvelle especed'Araneides appartenantau genreScytodes. 
Ann. Soc. Ent. France, 3, p. 73, pi. 1. 

Marx, George 

1889. A Contribution to the knowledge of the Spider Fauna of the 
Bermuda Islands. Proc. Phila. Acad., pp. 98-101, pi. 4. 

1890. Catalogue of the Described Araneae of Temperate North America- 
Proc. U. S. Nat. Mus., 12, pp. 497-594. 

Petrunkevitch, Alexander 

1925. Arachnida from Panama. Trans. Conn. Acad., 27, pp. 51-248, 
figs. 1-157. 

1926. Spiders from the Virgin Islands. Ihid., 28, pp. 21-78, figs. 1-28. 
1929-1930. The Spiders from Porto Rico. Ihid., pt. 1, 30, pp. 1-158, 150 

figs.; pt. 2, pp. 159-355, 240 figs.; pt. 3, 1930, 31, pp. 1-191, 168 
figs. 

Simon, Eugene 

1891-1897. On the Spiders of the Island of St. Vincent, pt. 1, Proc. 
Zool. Soc. London, pp. 549-575, pi. 42; pt. 2, ihid., 1894, pp. 519- 
526, 4 figs.; pt. 3, ihid., 1897, pp. 860-890. 



BRYANT: SPIDERS OF THE VIRGIN ISLANDS 363 

1892-1903. Histoire Naturelle des Araignees. 2nd. edition. Paris. 2 
volumes. Vol. 1, 1892, pp. i-vii, 1-1084, 1098 figs.; Vol. 2, fasc. 1, 
1897, pp. 1-192, figs. 1-200. fasc. 2, pp. 193-380, figs. 201-384; 
fasc. 3, 1901, pp. 381-668, figs. 385-792; fasc. 4, 1903, pp. 669- 
1081, figs. 793-1122. 

Taczanowski, Ladislat: 

1872. Les Araneides de la Guyane frangaise. Horae Soc. Ent. Ross., 9, 
pp. 64-150, pis. 3-6. 

1873. (continuation). Ibid. 10, pp. 56-115, pi. 2. 

Vinson, A. 

1863. Araneides des lies de la Reunion et Madagascar. 

Walckenaer, Charles Athan, Baron de. 

1802. Faune Parisienne, 2 volumes, Paris. Aranea, vol. 2, pp. 187-250. 
1837-1847. Histoire Naturelle des Insects Apteres. 4 volumes. Paris, 
Atlas \\ith 52 pis. 



EXPLANATION OF PLATES 



PLATE 1 



Bhvam'^X irgiii Islands Spiders 



PLATE 1 

Fig. 1 Dysderina antillana spec, nov., male, cephalothorax. 

Fig. 2. Oonopinus pretiosus spec, nov., female, sternum. 

Fig. 3. Telchius placidus spec, nov., male, cephalothorax. 

Fig. 4. Telchius placidus spec, nov., left palpus, retrolateral view. 

Fig. 5. Stenoonops nitens spec, nov., left palpus, retrolateral view. 

Fig. 6. Scaphioides reducta spec. nov.. epigynum. 

Fig. 7. Dysderina antillana spec, nov., left palpus, ventral view. 

Fig. 8. Scaphioides reducta spec, nov., female, cephalothorax. 

Fig. 9. Oonopinus pretiosus spec, nov., female, cephalothorax. 

Fig. 10. Stenoonops nitens spec, nov., male, cephalothorax. 

Fig. 11. Gamasomorpha perplexa spec, nov., male, cephalothorax. 

Fig. 12. Gamasomorpha perplexa spec, nov., left palpus, ventral view. 

Fig. 13. Hytanis pusilla spec, nov., female, cephalothorax. 

Fig. 14. Hytanis pusilla spec, nov., female, venter. 



BULL. MUS. COMP. ZOOL. 



Bryant: Virgin Island Spiders. Plate 1 




PLATE 2 



Bryant — Virgin Islands Spiders 



PLATE 2 

Fig. 15. Conopistha nephilae (Tacz.), left palpus, ventral view. 

Fig. 16. Lycosa subtilis spec, nov., left palpus, ventral view. 

Fig. 17. Caponina blanda spec, nov., left palpus, retrolateral view. 

Fig. 18. Gamasomorpha perplexa spec, nov., left palpus, retrolateral view. 

Fig. 19. Lycosa subtilis spec, nov., epigynum. 

Fig. 20. Thallumetus parvulus spec, nov., left palpus, retrolateral view. 

Fig. 21. Conopistha Cambridge] (Keys.), male, eyes, lateral view. 

Fig. 22., Thallumetus parvulus spec, nov., female, spinnerets. 

Fig. 23. Thallumetus parvulus spec, nov., male, eyes and mandibles. 

Fig. 24. Conopistha cambridgei (Keys.), left palpus, ventral view. 

Fig. 25. Lycosa reducta spec, nov., epigynum. 

Fig. 26. Conopistha nephilae (Tacz.), male, eyes, lateral view. 



BULL. MUS. COMP. ZOOL. 



Bryant: Virgin Island Spiders. Plate 2 




PLATE 3 



Bryant — Virgin Islands Spiders 



PLATE 3 

Fig. 27. Theridion virginus spec, nov., left palpus, ventral view. 

Fig. 28. Theridion brevipalpus spec, nov., male, front view of eyes. 

Fig. 29. Selenops lindborgi Petr., left palpus, ventral view. 

Fig. 30. Menemerus kochi spec, nov., epigynum. 

Fig. 31. Theridion brevipalpus spec, nov., male, right I femur, ventral view. 

Fig. 32. Theridion brevipalpus spec, nov., left palus, ventral view. 

Fig. 33. Corinna antillana spec, nov., epigynum. 

Fig. 34. Sidusa pavida spec, nov., left palpus, dorsal view of tibial apophysis. 

Fig. 35. Habronattus brunneus var. insgnis var. nov., epigynum. 

Fig. 36. Theridion guanicae Petr., left palpus, ventral view. 

Fig. 37. Selenops lindborgi Petr., epigynum. 

Fig. 38. Habronattus brunneus var. insignis var. nov., left palpus, ventral 

view. 
Fig. 39. Theridion cybele spec, nov., epigynum. 
Fig. 40. Sidusa pavida spec. nov,. left palpus, ventral view. 



BULL. MUS. COMP. ZOOL. 



Bryant: Virgin Island Spiders. Plate 3 







V 



Bulletin of the Museum of Comparative Zoology 

AT HARVARD COLLEGE 

Vol. LXXXIX, No. 8 



THE ECHINODERM FAUNA OF BERMUDA 



By Hlbert Lyman Clakk 



With One Plate 



CAMBRIDGE, MASS., U.S.A. 

PRINTED FOR THE MUSEUM 

April, 1942 



^f*^ Zoology ' 

APR 22 1942 

J^ < B R A Kl, 



No. 8. — The Echinodenn Fauna of Bermuda 
By Hubert Lyman Clark 

In 1888 appeared the first list of Bermudan echinoderms, that of 
Professor Angelo Heilprin of the Philadelphia Academy of Natural 
Sciences, who took a party of students tO the islands in the summer of 
that year. Their collections contained 6 nominal kinds of holothurians 
(now believed to represent only 4 valid species), 2 sea-stars, 6 brittle- 
stars and 6 sea-urchins, a total of 20 species. The list was published in 
December, 1888, in the Proceedings of the Academy of Natural 
Sciences. In September, 1898, 1 published "Notes on the Echinoderms 
of Bermuda," based on collections made by Professor C. L. Bristol of 
New York University in the summer of 1897. This appeared in the 
Annals of the New York Academy of Sciences (11, pp. 407-413) and 
includes 28 species but at least 5 of the holothurians were of doubtful 
validity. Further collections by Professor Bristol's students in 1898 
led to my publishing a revised list in July, 1899 (see Annals N. Y. Acad. 
Sci., 12, pp. 117-138), containing 29 species, but one of these is a 
synonym, one is only the young of another and a third is incorrectly 
identified. In April, 1899, I spent two weeks in Bermuda, and in June 
and July, another party from New York University was in the field. 
This field work led to my publishing a third paper entitled "Bermudan 
Echinoderms" (see Proc. Boston Soc. Nat. Hist., 29, pp. 339-344, 
May, 1901) which also includes 29 species and corrects errors in the 
earlier lists. 

Meanwhile Professor A. E. Verrill had begun his series of papers on 
the Bermudas and their fauna, which added much to our knowledge 
of the echinoderms. His first contribution appeared in the Trans- 
actions of the Connecticut Academy of Arts and Sciences, 10, pp. 583- 
587, in the fall of 1900 (the signatures are dated both September and 
October, though the cover says September) and is called "Additions 
to the Echinoderms of the Bermudas", using my first list as the basis 
for the additions. These were chiefly brittle-stars, several of which 
were not taken by the Yale party, but rest on earlier and dubious 
records. The "Synapta viridis" listed is synonymous with Synaptula 
hydriformis. In October, 1901, in the same journal (11, pp. 35-37) a 
second contribution appeared, called "Additions to the Fauna of the 
Bermudas", which lists 5 species of echinoderms, taken by the Yale 
Expedition in the spring and early summer of 1901. Only one of these 
species is an addition to previous lists. In April 1907, further notes on 
the echinoderms of Bermuda with special reference to habits and 



368 bulletin: museum of comparative zoology 

habitats appeared in volume XII of the Transactions. On pp. 100- 
102 are references to the various holothurians and echini which secure 
their nourishment by passing mud and sand through the ahmentary 
canal. A sea-star and a brittle-star are also discussed in connection 
with their sandy habitat. On pp. 275-285, a list is given of the Echin- 
oderms found on or about corals and the coral reefs, with notes as to 
their occurrence and habits. Altogether in these lists of 1907, Verrill 
reports a total of 36 species of Echinoderms in the Bermudan fauna. 

In 1919, in a paper on the Distribution of the Littoral Echinoderms 
of the West Indies (Carnegie Inst., Publ. 281, pp. 49-74), I listed 42 
species from Bermuda, but 2 or 3 of these are unquestionably mistakes. 
In 1922, my "Echinoderms from the Challenger Bank, Bermuda" 
(Proc. Amer. Acad., 57, pp. 353-361, pi. 1) appeared, listing 13 species, 
but only 8 of these are as yet known from Bermuda. In 1933, my 
"Handbook of the Littoral Echinoderms of Porto Rico and the other 
West Indian Islands" was published by the New York Academy of 
Sciences (Sci. Survey of Porto Rico, 16, pt. 1), with a list of 41 Bermu- 
dan species on p. 124. This however includes as an inexcusable error, 
a mythical species "Ophiozona imbricata" ; presumably Ophiozona im- 
pressa, long ago listed from Bermuda but apparently by error, as it has 
not been taken in the past half century. 

In April, 1939, 1 enjoyed a stay of 18 days at the Bermuda Biological 
Station, during which I did shore collecting at ten different localities, 
most of which were new to me. There was also one brief dredging trip 
in Ferry Reach. The best collecting for echinoderms I found at 
Hungry Bay, on the south shore, where 20 species were taken during 
my three visits. Only 8 additional species were secured at all of the 
other places visited. I am under great obligations to Dr. J. F. G. 
Wheeler for the hospitality of the Station and for his constant readi- 
ness to assist me in every way he could. Particularly I have to thank 
him for calling my attention to the strange occurrence of Synaptula 
hydriformis in Lovers Lake; for the opportunity to measure and ex- 
amine carefully a Bermudan specimen of Mellita quinqniesperforata; 
and for his generous gift of a Bermudan Oreaster, the only one I have 
ever seen. Dr. H. B. Moore was also assiduous in aiding me, particu- 
larly in getting living specimens of Leodia sexiesperforata, and Eucidaris 
tribvloides. In the summer of 1941, Dr. Moore kindly sent me semi- 
fossil remains of several species of echini secured by the dredging opera- 
tions in Castle Harbor. Among these were two recognizable tests of 
the spatangoid Moira atropos, not previously known from Bermuda. 
To Dr. E. F. Thompson and Dr. Marie V. Lebour, I am much in- 



CLARK: ECHINODERM FAUNA OF BERMUDA 369 

debted for encouragement and help. I wish to acknowledge also the 
great kindness of Mr. L. S. Mowbray of the Bermuda Aquarium at 
Flatts in making a day's collecting at Gravelly Bay and in Harrington 
Sound possible and successful. He furthermore showed me some im- 
portant specimens of Echini in the new Museum. It is a pleasure to 
give my hearty thanks to all of these friends who did so much to make 
our Bermudan visit of 1939 as profitable as pleasant. 

In presenting the following revised list of the Echinoderms of Ber- 
muda, it may be well to mention first those species previously listed 
whose names wall not be found in the following pages, and explain the 
cause of their absence. There are 9 of these, as follows: 

Nemaster iowensis — erroneously listed from "Bermuda" instead of 

"Bahamas." 
Ophidiasler guildingii — erroneously listed instead of Linckia guildingii. 
Ophiactis mulleri — listed by 'S'errill for mulleri Ltk. but there is no record 

of that species from Bermuda. 
Ophiopsila riisei — no reliable record from Bermuda. 
Ophiura hrevispina — no reliable record from Bermuda. 
Ophiozona imbricata — lapsus calami for 0. impressa. 
Ophiozona impressa — no reliable record from Bermuda. 
Holothuria floridana — no reliable record from Bermuda. Heilprin's record 

apparently refers to H. surinamensis. 
Leptosynapta inhaerens — mistaken identification. 

The following 36 names (arranged alphabetically) occur in one or 
more of the earlier lists but are now considered synonyms of more cor- 
rectly applied names. 

Actinopyga parvula = Holothuria parvula 

Amphipholis goesi = Amphipholis gracillima 

Amphipholis tenera = Amphipholis squamata 

Asterias atlantica = Stolasterias tenuispina 

Asterias tenuispina = Stolasterias tenuispina 

Astroporpa affinis = Asteroporpa annulata 

Chondrocloea vivipara = Synaptula hydriformis 

Cidaris tribuloides = Eucidaris tribuloides 
Coscinasterias tenuispina = Stolasterias tenuispina 

Cucumaria punctata = Thyone surinamensis 

Diadema setosum = Centr echinus antillarum 

Echinoneus semilunaris = Echinoneus cyclostcmus 

Hipponce escidenta = Tripneustes esculentus 

Holothuria abbreviata = Holothuria parvula 

Holothuria captiva = Holothuria parvula 



370 



bulletin: museum of comparative zoology 



Holothuria rathbuni 
Leptosynapta acanthia 
Leptosynapta roseola 
Mellita sexforis 
Mellita sexiesperforata 
Ophiactis krtbsii 
Ophiactis lymani 
Ophiocoma crassispina 
Ophiura nppressa 
Ophiura brevicauda . 
Ophiura cinerea 
Semperia bermudensis 
Stichopus diaboli, 
Stichopus moebii 
Stichopus xanthomela 
Synapta acanthia 
Synapta iiihaerens 
Synapta roseola 
Synapta viridis 
Synapta vivipara 
Toxopneustes rariegatus 



= Holothuria arenicola 

= Eupatinapta acanthia 

= Epitomapta roseola 

= Leodia sexiesperforata 

= Leodia sexiesperforata 

= Ophiactis saiignyi 

= Ophiactis algicola 

= Ophiocoma echinata 

= Ophiodenna appressuni 

— Ophioderma brevicaudum 

= Ophioderma dnereum, 

= Thyone snrinamensis 

= Stichopus badionotus forma diaboli 

= Stichopus badionotus 

= Stichopus badionotus forma xanthomela 

= Eupatinapta acanthia 

= Leptosynapta ijihaerens (non-Bermudan) 

= Epitomapta roseola 

= Synaptula hydriform,is 

= Synaptula hydriformis 

= Lytechiuus variegatus 



Making due allowance then for these errors and synonyms, 48 
species of Echinoderms are now known from Bermuda itself, not in- 
cluding Challenger Bank. This group of 48 is made up of 5 species of 
sea-stars, 18 brittle-stars, 12 echini and 13 holothurians. Artificial 
keys to aid in identifying these species will be found in the "Handbook 
of the Littoral Echinoderms of Porto Rico, etc." mentioned above 
(p. 3, lines 11-14), excepting only one or two deep water forms. 



CRINOIDEA 

I'eather-stars 

In 1907, Verrill recorded "a single specimen" of Antedon "too young 
for accurate identification" which "was obtained in 1901". This is 
apparently the only crinoid that has as yet been found in the Bermu- 
das. In 1919, I recorded Nem^aster ioicensis (Springer) from Bermuda 
on the strength of Mr. A. H. Clark's statement that he had a specimen 
from there. Subsequently Mr. Clark discovered that his specimen was 
from the Bahamas and not from Bermuda. Ob^'iously a crinoid fauna 
is practically wanting in the Bermudas. 



CLARK: ECHINODERM FAUNA OF BERMUDA 371 

ASTEROIDEA 

Sea-stars 

The scarcity of sea-stars in the marine fauna of Bermuda is one of 
its striking features. While 5 species are now known to occur, only one, 
Stolasterias, is at all common. Just why this should be so is not easy 
to understand, but it may be remarked that the entire West Indian 
littoral fauna contains but 17 species of sea-stars as compared with 40 
in the Torres Strait region. The fauna from which Bermuda has de- 
rived her sea-stars is thus itself somewhat depauperate. It is hard to 
see, nevertheless, why Astropecten and Echinaster should not occur, 
since they are common on the shores of the southeastern United States 
and in the Bahama Islands. On the other hand, Stolasterias which 
occurs commonly at Bermuda, is unknown on the coast of the United 
States and is \'irtually unknown in the West Indies; a record from 
Cuba needs confirmation. It is common in the Mediterranean and the 
eastern Atlantic and it seems very probable that it reached Bermuda on 
foul ship bottoms. 

1. LuiDIA CLATHRATA (Say) 

This sea-star lives on sandy bottoms, where it may be more or less 
completely buried beneath the surface. It is not strange therefore 
that it was not taken until 1898, when a party from New York Univer- 
sity secured a specimen in Harrington Sound; the color of this individ- 
- ual was said to have been "salmon-pink" in life. The following year 
another party from the same University was located on Whites Island 
in Hamilton Harbor and secured "several specimens on the beach", 
but these were blue-gray above and cream-color beneath — the usual 
coloration of specimens from the Carolinas, Florida and the W^est 
Indies. 

In 1901 \'errill reported: "Several fine specimens of this species were 
taken on a white shell-sand bottom in shallow water, at Trunk Island, 
Harrington Sound. It also occurred at Long Bird Island and other 
localities on shell-sand bottoms in shallow water. Its presence is indi- 
cated by a star-shaped impression in the sand. But it moves about 
under the sand with remarkable rapidity, when disturbed, by means of 
its large ambulacral tubes, so that it is not easy to capture it after it 
has taken alarm. Its color in life is generally light cream-color often 
with a rosy or flesh-colored tint and frequently with a darker grayish 



372 bulletin: museum of comparative zoology 

or greenish median streak on each ray. It becomes at least a foot in 
diameter at Bermuda." 

In 19C6, Verrill again refers to Luidia as "common in some places, 
living under the surface of the sand. It is remarkable for the rapidity 
with which it can glide along, using its numerous large ambulacral feet 
as paddles or oars for swimming or gliding, while concealed just under 
the loose sand." In 1915, Verrill says of this Luidia: "Bermuda, com- 
mon in sheltered situations just below low-tide level on bottoms of 
shell-sand, concealed just below the surface. All living specimens seen 
by me in Bermuda were either pale salmon or rose-salmon. Those from 
Florida and North Carolina are gray." He further repeats his earlier 
statements regarding the ease and rapidity of its movements and 
asserts further "It can also swim . . . free of the sand or on its sur- 
face" by means of "its large flattened muscular feet" which it "uses 
like paddles". 

I have not been so fortunate as to meet with this sea-star alive in 
Bermuda, but have collected it on the Carolina coast and in Jamaica. 
The specimen taken by the New York University party in Harrington 
Sound, Bermuda, in 1898 is one of about 30 specimens in the M. C. Z. 
collection : it is a dull cream-color and has R= 140 mm. AH very young 
individuals in the M. C. Z. collection are very light colored, practically 
without pigment. As size increases, pigment usually appears in the 
median part of the upper surface of each ray. As a rule it spreads over 
the whole abactinal surface, though deepest and darkest along the 
radial median area. In a few individuals, the pigment is confined to 
that area so definitely, that the dark longitudinal line on each ray 
stands out conspicuously against the unpigmented background. Pink 
or salmon tints have not been reported from any place but Bermuda 
and further, more detailed observations on this point should be made. 
Even more important are careful observations on the movements and 
possible swimming powers of Bermudan Luidias, as reported by Ver- 
rill. No living specimen seen by me has shown any activity whatever. 



2. Oreaster reticulatus var. bermudensis, var. nov. 

Figures 1 and 2 

The occurrence of Oreaster at the Bermudas has never been defin- 
itely recorded, but a specimen presented by Dr. J. F. G. Wheeler to the 
M. C. Z. in April, 1939, establishes it beyond dispute. This specimen 
was detected with a water glass on the bottom of Ferry Reach and was 



CLARK: ECHINODERM FAUNA OF BERMUDA 373 

then brought up by a diver and given to Dr. Wheeler. It is very large 
and obviously senescent, but it is so strikingly different from any West 
Indian specimen of Oreaster seen, that it seems proper to give it a 
varietal name. It is now dry and shows the effects of handling. All 
the rays are turned upward at the tip but there is no doubt that in 
life R exceeded 170 mm. and r is even now nearly 90. The rays are 
very wide at base, 90 mm. or more, blunt and rounded at the tip; 20 
mm. orad, the ray is 35 mm. wide but not much over 20 mm. high. 
Vertical diameter at center of disk is about 60 mm. Abactinal skeleton 
most irregularly and imperfectly reticulated; there are no regular 
series of tubercles and the papular areas are large, irregular and ill- 
defined. The central portion of the disk, a space 60-70 mm. in diame- 
ter, is rather definitely delimited by narrow skeletal plates bearing 
conical tubercles, 2-4 mm. high and 2-3 mm. in diameter at base; on 
these tubercles most of the cone is bare and smooth but the basal part 
is finely and evenly granulated. There are about 15 supero-marginal 
plates on each side of each ray but at the tip they are abruptly smaller, 
crowded and ill-defined. Each normal plate carries a conspicuous 
tubercle about 8 mm. high (or long), the terminal portion (5 mm. long 
by 4 mm. in diameter) bare and smooth; occasionally double, the tip 
is more or less pointed, but the point is often irregular and may be 
flattened laterally to a greater or less degree. Near the end of each 
ray, the tubercles on all plates are much smaller and more irregularly 
placed. 

Inferomarginal plates, entirely on the oral surface, overshadowed 
by the superomarginals, which they resemble more or less but they 
often have 2, or even 3, cones and the granulation is distinctly coarser, 
at least on the orad side. The rest of the oral surface is covered by 
convex, granulated actinal intermediate plates, each of which carries 
one, or very often 2 smooth cones, 2.5-5 mm. high and 2-4 mm. thick 
at base. Granulation of these plates is largely made up of pedicellariae 
and since, in this specimen, the jaws of these pedicellariae have for the 
most part fallen out of their sockets, the granulation appears to be 
made up for the most part of deeply excavated circular or elliptical 
cups. Adambulacral armature heavy, a single series of 5 or 6 unequal, 
flat, truncate furrow spines, back of which a very large irregularly 
conical spine (or often two of them) covers the actinal surface of the 
plate; these spines are commonly more or less flattened parallel with 
the furrow and show great diversity of size and form. Oral armature 
made up of similar but blunter spines; it is too deeply sunken however 
to permit a description of its details. 



374 bulletin: museum of comparative zoology 

Upper surface light reddish-brown, the large tubercles and tips of the 
arms lighter, often nearly white. The lower surface has evidently been 
bleached to some extent; the many bare plates are light reddish-brown 
while the spines and granules are pure white, as are many large actinal 
plates, which have lost their epidermis. 

This peculiar Oreaster differs from all the specimens of reticulatus 
in the M. C. Z. in the short blunt rays, the entire lack of serial arrange- 
ment of spines or papulae on the upper surface and the large conical, 
usually single, spines on the actinal plates. There are several West 
Indian specimens at hand larger than this one, and at least one is as 
obviously senescent, but they do not look at all like the Bermudan 
specimen. For the present therefore it may well bear the varietal name 
suggested. Obviously further material is much to be desired. 

3. LiNCKiA GUiLDiNGii Gray 

(= OphJdiasfrr (juildingii Verrill, 1900, p. 584, xo\ Ophidiaster 
guildivgii Gray, 1840.) 

This stiff and inert species has been taken many times in Bermuda, 
but it is not common anywhere. Adults are very rarely found, most of 
the individuals taken having the rays less than 100 mm. long. Small 
specimens often have 6 or rarely 7 rays but in adults 5 is the normal 
number. The species is tropicopolitan but the largest specimen yet 
recorded (M. C. Z. no. 2648) was taken July 2, 1903 at Bermuda; its 
longest ray is about 215 mm. The color of this sea-star shows ex- 
traordinary diversity as violet, bright blue, dull red variegated with 
lighter, and yellow-brown specimens are recorded. In Bermuda, how- 
ever, the color seems to be always dull orange or yellow-brown. Large 
individuals occur lying exposed on sandy bottoms, but the great ma- 
jority of specimens are found under rocks or concealed in crannies in 
the reef. 

4. AsTERiNA folium (Liitken) 

Living on the under surface of rock-fragments near low water mark, 
this sea-star is easily' overlooked. It is flat, pentagonal and rarely 
over 20 mm. across. Its color when adult is bluish-green or quite blue, 
of a light or dull shade. Young specimens may be cream-color or 
yellowish but they begin early to accumulate pigment dorsally. 
Asterina has been taken at Coney Island and in Castle Harbor, but 
apparentl^^ it is not very common. 



CLARK: ECHINODERM FAUNA OF BERMUDA 375 



5. Stolasterias tenuispina (Lamarck) 

Although an eastern Atlantic and Mediterranean species, this 
Stolasterias, easily recognizable as a rule by having 6-8 unequal arms, 
is the one common sea-star in Bermuda. It probably came to the 
islands originally on foul ship bottoms. It is now common and widely 
distributed and grows to a fairly large size, but specimens with R 
exceeding 90 mm. are seldom seen. In x\pril, 1939, Clark G. 
Myers collected a very symmetrical 7-armed specimen with R = 75- 
100 mm. and later, in Harrington Sound, he took a less symmetrical 
8-armed specimen, with R ranging from 60 to 110 mm. 

The largest specimen in the M. C. Z. collection (no. 1369) is an 
old one, labelled as from Bermuda. It is symmetrically 5 rayed and 
R= 135 mm. The history of this specimen is of more than ordinary 
interest, for it was collected by A. S. Bickmore of the Museum staff, 
when accompanying the celebrated P. T. Barnum on an expedition 
to the Bermudas in 1862. Professor Louis Agassiz, in his report of 
February, 1863, says: "We are indebted to Mr. P. T. Barnum for the 
facilities he has given to Mr. Bickmore of accompanying his expedition 
to Bermuda, during which Mr. B. has made an extensive collection of 
the marine animals of that island, showing that its fauna is identical 
with that of the coast of Florida." 



OPHIUROIDEA 

Brittle-stars 

Although more than a third of the species of Bermudan echinoderms 
are brittle-stars, only a single species, Ophionereis reticulata, is so 
widely distributed and so common that it will be found by the casual 
collector. It was only on the fifth day of diligent search at or near 
low-tide level, in April, 1939, that we found a second species, 
Ophiomyxa flaccida, and we did not see it again in the following two 
weeks! On the south shore, at Gravelly Bay and Hungry Bay, the 3 
species of Ophiocoma are fairly plentiful and Ophioderma appressum 
is not hard to find in Harrington Sound. The remaining dozen species 
are either small, secretive forms, seldom noticed unless search is being 
made for them, or larger species associated with a particular habitat, 
such as gorgonians or sponges. The handsome colors of Ophiomyxa 
flaccida and the bright rusty-red lower surface of Ophiocoma riisei 



376 bulletin: museum of comparative zoology 

make those species attractive objects, even to a person who is not a 
zoologist, but the other Bermudan brittle-stars are not conspicuous 
animals. 

1. Ophiomyxa flaccida (Say) 

This smooth, bright-colored, large and active brittle-star is found 
more or less frequently under rocks at or below low water mark near 
the Causeway, Long Bird Island and Coney Island. It was not found 
at Hungry Bay, and is not yet recorded from the southern or western 
shores of Bermuda. It is a typical West Indian species ranging from 
Bermuda to Brazil. 

2. AsTEROPORPA annulata Liitken 

Verrill (1901) records (under the name Astroporpa affinis Liitken) 
4 specimens of this small-bodied, long-armed ophiuran from "a large 
gorgonian {Vcrnicclla grandis Verrill) brought up from about 100 feet, 
off the outer reefs, on a fisherman's hook". It probably occurs where- 
ever this gorgonian (now Scirpearia grandis (Verrill)) or similar forms 
are growing, and it may occasionally be washed up on a beach, but 
it is unlikely that it occurs normally in less than 10 fms. of water and 
hence it is not included among the strictly littoral echinoderms of the 
West Indian region. 



-o' 



3. Amphipholis gracillima (Stimpson) 

Verrill (1900) reports this long-armed, small-bodied amphiurid, 
under the name "Aviphipholis goesi Ljung.", as having been dredged 
in 20-30 ft. of water in Baileys Bay in 1898. The species is known from 
Charleston, S. C, through the West Indian region to Rio de Janeiro, 
Brazil, but I have seen no specimens from Bermuda. At Sandy Point, 
Buccoo Bay, Tobago, it is quite common, buried deeply in the sand, 
just below low-water mark. 

4. Amphipholis squamata (Delle Chiaje) 

This almost cosmopolitan amphiurid is common in Bermuda, but 
owing to its small size (usually less than 4 mm. across the disk) and 
secretive habits (hidden in shells or rock crannies) it is very easily 
overlooked. Verrill (1900) lists it under the name Amphipholis tenera 
(Ltk.) Ljung. 



CLARK: ECHINODERM FAUNA OF BERMUDA 377 

5. Ophiostigma isacanthum (Say) 

This small West Indian brittle-star seems to be very rare in Ber- 
muda. Apparently neither Verrill nor the New York University parties 
ever met with it, nor have I. The Heilprin party (1888) took "two 
very young specimens, dredged in Harrington Sound". In July 1917, 
Dr. W. J. Crozier took 2 specimens in Fairyland Creek, a large bay in 
the western coast of Pembroke; these specimens are now in the 
M. C. Z. collection: one is a normal but rather small 5-armed adult, 
3 mm. across the disk; the other is a 6-armed specimen little more than 
2 mm. across the disk with 3 arms of one side smaller than the opposite 
3, indicative of reproduction by schizogenesis. 

6. Amphiodia repens (Lyman) 

This is another small amphiurid, which is probably fairly common, 
but is seldom noted, owing to its small size and secretive habits. It 
is rarely as much as 4 mm. across the disk, but the slender arms may be 
8 times as much. The disk is light gray but the arms are white in 
rather marked contrast : frequently, however, there are gray markings 
on the arms and especially on the surface of many of the arm-spines; 
these dark spots on some or many arm-spines may be on either the 
upper or under surface or on both. This brittle-star lives more or less 
buried in undisturbed sand, usually where some eel-grass or similar 
vegetation is growing, and may be discovered by sifting such sand 
through a sieve. In the cove at Coney Island where Eupatinapta 
acanthia is found, repens is frequently met with. 

7, Ophiactis algicola H. L. C. 

This is still another secretive little brittle-star, but its habits are 
quite unlike the preceding. It lives among coralline algae, bryozoa, 
etc. on rock fragments and never buried in sand. The arms are rela- 
tively short and like the disk are variegated with light and dark shades 
of gray and brown. The only specimen as yet known from Bermuda 
was taken June 24, 1916, at "Brackish Pond Flat", by Dr. W. J. 
Crozier, and is now in the M. C. Z. collection. 

8. Ophiactis savignyi (Miiller and Troschel) 

This very common tropicopolitan species may be found almost any- 
where in the shallow waters of Bermuda living "in the interstices of 



378 bulletin: museum of comparative zoology 

sponges and corals, often gregariously while young" (Verrill, 1900, p. 
586, under the name Ophiactis Krebsii Liitken). Adults are rare, 
nearly all specimens taken having 6 arms, and the disk less than 5 mm. 
across. 

9. Ophiothrix angulata (Say) 

Verrill (1900, p. 585) says this species is "not common" but he gives 
no single record of its occurrence in the Bermudas. It does occur at 
Challenger Bank, but its presence in the Bermuda group itself needs 
confirmation. 

10. Ophiothrix suensonii Liitken 

Verrill (1900, p. 585) includes this species as a Bermudan brittle- 
star but rests its claim on the statement that it was "collected at Ber- 
muda by Mr. G. Brown Goode". Its occurrence at the Challenger 
Bank is well attested, and it is quite probable that it lives among the 
Gorgonians of the deeper waters of Bermuda itself. 

11. Ophionereis reticulata (Say) 

This is probably the commonest Bermudan brittle-star, as it occurs 
wherever rock fragments rest on clean sandy bottoms. Large speci- 
mens often show a reddish tinge and in some cases the net-work of 
lines on the disk is quite orange-red. These brighter shades disappear, 
as a rule, in preserved material. 

12. Ophiocoma echinata (Lamarck) 

Common under rocks near low water mark on sandy bottoms. 
Reaches a large size (30 mm. across disk) at Gravelly Bay and Hungry 
Bay on the south shore. 

13. Ophiocoma pumila Liitken 

Common but much more secretive than the preceding species. 
Adults manage to conceal themselves among algae or in crannies in 
rocks and corals where their dull-brown color helps to hide them. The 
6-armed, green and white young are more conspicuous and much more 
easily found; their resemblance to Ophiactis savignyi is great enough to 
make confusion between the two species easy. 



CLARK: ECHINODERM FAUNA OF BERMUDA 379 

14. Ophiocoma riisei Liitken 

Common at all suitable localities, such as Gravelly Bay and Hungry 
Bay, where it reaches maximum size (disk, 30-35 mm. in diameter, 
in life). The rusty-red color of the oral surface is very striking, and 
as it is constant, it makes a convenient "recognition mark" in the field. 
Unfortunately it is less obvious in preserved material. 

Note: Verrill (1900, p. 586) reports Ophiopsila riisei Liitken from 
Bermuda, giving Lyman as his authority. There is some mistake about 
this, for there are no Ophiopsilas from Bermuda in the M. C. Z., and 
it is quite unlikely that Lyman ever had any specimens from there. It 
is possible that Lyman may have written "Bermuda" at some time 
when he intended to have written "Bahamas," but this is merely a 
guess. Verrill's later listing (1907, p. 284-328) is not substantiated by 
any definite evidence. 

15. Ophioderma appressum (Say) 

Fairly common on sandy bottom near low water mark, usually under 
rocks. 

16. Ophioderma brevicaudum Liitken 

Reported only by Verrill (1903, p. 584) who says it was "not com- 
mon" in 1898. I have never seen a Bermudan specimen. In 1907 
(p. 282 (= 326) ) Verrill also lists Ophioderma bremspinum (Say), but 
gives no authority for its occurrence in Bermuda. I have never seen 
or heard of a Bermudan specimen. 

17. Ophioderma cinereum Miiller and Troschel 

This is one of the larger West Indian species, reaching a disk diame- 
ter of 30-35 mm. in Tobago where it is very common. Verrill says it 
was taken in Bermuda by the Yale party in 1898, but as he says it 
occurs in "interstices and crevices of the reefs," I am dubious as to 
the identification of his material. In Jamaica and Tobago, it occurs 
only under slabs of broken coral and rock, and makes no attempt to get 
into interstices and crevices. Its large size and rather rigid arms are 
quite out of keeping with hiding in crevices! 

18. Ophiolepis paucispina Miiller and Troschel 

This little brittle-star proves to be fairly common, after one has 
learned where and how to find it. It lives on or in the sand under rock 



380 bulletin: museum of compakative zoology 

fragments, and when such a fragment is turned over, it seems to be 
sucked up and forced to He with its aboral surface against the lower 
side of the rock. As the lower surface is white or cream-color and the 
arms are coiled close to its margin, the brittle-star becomes rather 
difficult to detect. At Hungry Bay, Ophiolepis is quite common, but 
it does not seem to be so numerous along the north coast, though we 
took specimens at the cove on Coney Island in 1939. 



ECHINOIDEA 

Echini, sea-urchins, key-hole urchins, heart-urchins, etc. 

Of the 12 Bermudan echini, only one, Lytechinus variegatus atlanticus, 
can be considered really common. In particular localities, Centrechi- 
nus, Tripneustes and Echinometra, can be secured in considerable 
numbers, under favorable conditions, and Leodia sexiesperforata is ap- 
parently fairly common on those pure sandy bottoms which it favors. 
The remaining half dozen species are distinctly rare or restricted to 
such inaccessible areas that they are rarely taken. Of two or three of 
these, living specimens have not yet been secured. 



1. EuciDARis TRiBULOiDES (Lamarck). 

Not very common, but occurs at North Rock, Coopers Island, and 
other similar reef areas. 



2. Centrechinus antillarum (Philippi) 

Fairly common on the outer reefs and on rough, rocky shores, as at 
Coopers Island, Hungry Bay, etc. Listed by Verrill as "Diadem,a 
sctosum". 



3. Lytechinus variegatus atlanticus (A. Ag.) 

Very common on sandy or "grassy" bottoms. Young ones are also 
to be found under or among rocks. Often the rich dark red-violet 
color is not acquired until the urchin is half-grown, young ones being 
greenish, with more or less white and even with pinkish tips to the 
primary spines, as in Florida specimens. 



CLARK: ECHINODERM FAUNA OF BERMUDA 381 



4. Tripneustes esculentus (Leske) 

Fairly common and reaching a large size, up to 150 mm. in diameter, 
the maximum for the species. Young ones are found under and among 
rocks but the adults occur out on the open bottom. Apparently there 
is an absence of enemies and conditions favor growth. 



5. Echinometra lucunter (Linn.) 

Very common on suitable rocky areas, and conditions apparently 
favor growth, as in the case of Tripneustes, for the largest known 
specimens (85-90 mm. in the greater diameter) were found in Bermuda. 



6. Clypeaster rosaceus (Linn.) 

This species has not hitherto been recorded from Bermuda, but there 
are bare tests in the Museum at the Aquarium which I was assured 
were taken locally. Unfortunately there are no data with the speci- 
mens. Similar bare tests were seen at a curio store in St. George but it 
is possible they came from further south. 



7. Clypeaster subdepressus (Gray) 

This species, like the preceding, has hitherto been unrecorded from 
Bermuda, but there is at least one bare test in the Museum at the 
Aquarium which I was assured was taken in Bermuda. The absence of 
definite data is greatly to be regretted. 



8. Mellita quinquiesperforata (Leske) 

Verrill (1907, pp. 144 and 188) lists this species (under the name 
testudinata) as having been recorded by R. J. Nelson in 1840 as a 
fossil occurring on "Ireland Island and the islands in Crow Lane, 
Hamilton". He adds: "It has not been observed here by others, either 
living or fossil". It is therefore a matter of very great interest that 
there is a specimen at the Bermuda Biological Station which Dr. 
Wheeler assures me was dredged in the vicinity of St. Georges, or at 
least in the northeastern waters of Bermuda. This specimen is bleached 
to a pure white and is fragile. It is 82 mm. long by 84 mm. wide, but 
as the posterior margin of the test is concave, the full length may be 



382 bulletin: museum of comparative zoology 

reckoned as 84 mm. The form is rounded pentagonal with all the 
angles rounded and all the sides concave, the posterior margin most 
markedly so. The posterior petals (I and V) are 14-15 mm. long and 
6 mm. wide while the anterior (II and IV) are 13-14 mm. long and 
5.5 mm. wide; unpaired petal (III) is 16 X 6 mm. All the petals are 
narrowly open. The posterior lunules (I and V) are 14 mm. long; V is 
about 1.5 mm. wide, but I is aberrant and somewhat deformed, 2 mm. 
wide at the proximal end and 5 mm. wide distally. Anterior lunules II 
and IV are 1 1 by 1 mm. and the posterior interradial lunule is 12 by 2. 
Owing to the way in which the specimen is mounted for exhibition, the 
oral surface could not be examined. 

In all of the more than 800 specimens of this Mellita in the M. C. Z. 
as well as among the scores I have examined elsewhere, I have never 
seen a specimen with the sides so markedly concave as in this individual 
from Bermuda. Another remarkable feature is seen in the small 
petaloid area; in a typical specimen from Florida, 80 X 83 mm. the 
petaloid area is about 50 mm. long by 45 wide, but in this Bermudan 
specimen, of practically the same size it is hardly more than 35 by 32. 
In the Florida specimen, petal III is 21 mm. long, in the Bermudan it 
is but 16, and the other petals are proportionate. A third notable 
character of the Bermudan Mellita is the small size of the unpaired 
lunule which is only 12 mm. long while the posterior paired lunules 
are 14. In the specimen from Florida, the posterior paired lunules are 
also 14 mm. but the unpaired lunule is 19 mm. 

In view of these striking differences it is probable that the Bermudan 
Mellita is a local species distinct from qimtquicsperforata, or at least a 
well-marked variety. But until more material is available and we have 
some idea of the extent of individual diversity in the Bermudan form, 
it is hardly wise to designate it by a new name. 

9. Leodia sexiesperforata (Leske) 

The revival of Gray's (1851) generic name for the key-hole urchins 
with 6 lunules, which Lambert and Thiery proposed in 1921, seems 
quite justified since we now know that the lunules originate in a very 
different manner from that found in Mellita. In Leodia they arise by 
resorption, in Mellita by the closing distally of marginal notches, a 
most extraordinary difference. 

Leodia appears to be common at suitable places in Bermuda but 
exact localities have not been recorded. Dr. H. B. Moore took us, 
April 18, 1939, on an excursion to West Whale Bay, on the west coast 



CLARK: ECHINODERM FAUNA OF BERMUDA 383 

of Southampton parish, where there is a wide area of shoal water with a 
clean sand bottom in which Leodia seems to thrive. Of the specimens 
we secured, two were exceptionally large and one, taken by Clark G. 
Myers, proved to be the largest representative of the species yet re- 
corded, as it measured 117 by 125 mm. In life Leodia is a pale fawn 
color but this becomes a deep green in killing agents of any sort, even 
fresh water. If, however, specimens are dried very thoroughly, directly 
from the sea-water, the salt can later be soaked out of them without, 
as a rule, affecting the color. 

10. EcHiNONEUS CYCLOSTOMUS Leskc 

This sluggish tropicopolitan echinoid was long since recorded from 
Bermuda, but very few specimens have been taken. A. H. Verrill 
took two from sand under rocks, at very low tide in Hungry Bay in 
March, 1901, but diligent search by the writer during three visits to 
Hungry Bay in 1939, failed to disclose a single specimen. At the Ber- 
muda Biological Station, there is a fine bare test of large size (37 X 30 
mm.) which was found at Coopers Island. 

11. MoiRA ATROPOS (Lamarck) 

This unmistakable spatangoid, well known from the southeastern 
coast of the United States and the northern West Indies, has never 
been taken in Bermuda, except for the two semi-fossil specimens 
dredged in Castle Harbor, referred to on p. 368. These are of maximum 
size (60 mm. long) and heavily covered with deposited lime. Whether 
Moira is now extinct in Bermuda or still lives in suitable localities re- 
mains to be determined. 

12. Brissus brissus (Leske) 

This spatangoid, common to the Mediterranean and the West Indies, 
seems to be rare at Bermuda, but there are specimens in the Museum 
at the Aquarium which I was assured were Bermudan. In the West 
Indies, Brissus is usually small (50-70 mm. in length) but in the eastern 
Atlantic and Mediterranean it is nearly twice as large (115-135 mm.). 
The specimens at the Aquarium were larger than any Jamaican speci- 
mens I have seen. Like Echinoneus, with which it is frequently asso- 
ciated, Brissus lives buried in sand under rocks, and its natural color is 
very similar to that of the sand. 



384 bulletin: museum of comparative zoology 

HOLOTHURIOIDEA 

Holothurians, sea-cucumbers, synaptids, etc. 

Of the 13 species of holothurians, the big Stichopus is easily most 
abundant, and is moreover one of the commonest and most conspicuous 
of the marine invertebrates of Bermuda. Almost equally common but 
so much smaller and more secretive, that it is seldom noted by the 
average person, is Chiridota rotifera, found in the sand under rocks 
along shore wherever conditions are not unsuitable. Of 2 species 
{Actinopyga agassizii, Holothuria iinyaticns) only a single specimen of 
each has been taken as yet, and oddly enough both were found at the 
entrance to Hungry Bay. They are common West Indian forms and 
will probably be .found ultimately in Bermuda in reasonable numbers. 
It is not impossible however that these two species have arrived from 
the south as larvae or very young individuals on ship bottoms, and are 
not yet fully established. Of the 5 species of Holothuria, 2 (parvula and 
surinamensis) are cornmon and may be found on the under side of rock 
fragments near low watermark, wherever the water is clean and well 
aerated. Another species, arenicola, is common, at least on the south 
coast, deeply buried in sand under rock-fragments, but is easily over- 
looked. In a similar habitat occurs a fourth species {cuhana) which 
seems to be rare. There are only 2 dendrochirote species; one, a 
Thyone, seems to be common in some seasons or localities and rare at 
others; the second, a psolid, has only just been discovered at Hungry 
Bay, where it is not very rare but is hard to detect. There are 3 species 
of worm-like synaptids which are more or less common, but they are 
so concealed by either their coloring or their habits that they are seldom 
noted save by the collector who knows where to locate them. 

1. Holothuria cubana Ludwig 

This small whitish holothurian lives buried in sand close to or under 
rock-fragments. It is very sluggish and secretive. The body wall is 
firm and rather rough, in part from sand grains which adhere tightly 
to it. We found a single specimen in April, 1939, near the "Causeway", 
on the Ferry Reach side. 

2. Holothuria arenicola Semper 

This tropicopolitan species is common on the southern coast of 
Bermuda, particularly at Gravelly Bay and Hungry Bay. It grows to 



CLARK: ECHINODERM FAUNA OF BERMUDA 385 

a large size (250-300 mm. long), but always lives deeply buried in the 
muddy sand under rock-fragments. Owing to its dull sandy, or often 
rusty, color it is easily overlooked. There are always spots or blotches 
of dusky gray or dull purplish, but these show great diversity in num- 
ber, size and distribution. 

3. HoLOTHURiA iMPATiENS (Forskal) 

A very fine and typical representative of this tropicopolitan species 
was found under a rock near low water mark at Hungry Bay, April 24, 
1939. Its characteristic extrusion of copious Cuvier's organs was not- 
able. Apparently this is the first record of impatiens in Bermuda. 

4. HoLOTHURiA PARVULA (Sclcnka) 

This is a very common and generally distributed small holothurian, 
usually under 60 mm. in length and rarely, if ever, exceeding 75 mm. 
The bright or deep brown color, flattened ventral surface and copious 
Cuvier's organs, make its recognition easy. It occurs always on the 
under surface of rock fragments in shallow water, and there are fre- 
quently several on a single piece of rock — sometimes there are 10-12. 
In all previous Bermudan lists, this species has been called H. captiva 
but parvula is the earlier name. 

5. HOLOTHURIA SURINAMENSIS Ludwig 

This is also a very common holothurian on the lower surface of rock- 
fragments in shallow water. It is usually about 100 mm. long but 
large individuals are notably extensile and may stretch to about twice 
that length. It is often found with parvula, but is readily distinguished 
from that species by its more cylindrical form, whitish or light yellowish 
tentacles and absence of Cuvier's organs. The color ranges from a 
reddish-brown to a dull purplish-gray, with or without more or less 
blotching with a different shade. Some of the large gray specimens are 
superficially quite like impatiens but are readily distinguished by the 
pedicels and the absence of Cuvier's organs. 

6. AcTiNOPYGA AGASSizii (Sclenka) 

There is a single record of this species from Bermuda (W. J. Crozier, 
1917, Ann. Mag. Nat. Hist. (8) 19, p. 405) but unfortunately the speci- 
men was not preserved. It was an adult, about 250 mm. long, taken 



386 bulletin: museum of comparative zoology 

in some 6 feet of water at the entrance to Hungry Bay. Further col- 
lecting by various individuals, at Hungry Bay and elsewhere, has 
failed to produce another representative of the genus. There is no ap- 
parent reason why the species should not be a permanent member of 
the Bermudan fauna. 

7. Stichopus badionotus Selenka 

This common West Indian holothurian is one of the .most obvious 
and widely distributed of the Bermudan echinoderms. It seems to be 
present almost everywhere that the water is clear and well aerated. 
Unlike the badionotus of Jamaica, the Bermudan Stichopus is re- 
markably constant, in coloration, but in two forms. The great ma- 
jority of individuals are unicolor and commonly so dark they appear 
to be black. Close examination in good light shows that the lower sur- 
face is lighter than the upper, and the real color is a very deep purple or 
blackish-brown. In some cases the color is definitely brown or even 
yellowish-brown, but these lighter shades are not common. The other 
form is a light wood-brown, orange-brown or orange-buff, more or less 
blotched and spotted with black or black-brown. This maculated 
form looks so different from its unicolored congener, that Heilprin 
(1888) regarded them as distinct species, each of which he thought was 
undescribed. The unicolored one he called diaholi and the spotted one, 
xanthovuia. These words may be conveniently used as varietal or 
"form" names for Bermudan material, but in Jamaica no such distinc- 
tion is possible, as spotted and unicolor forms mingle in inextricable 
confusion. It has been suggested that the spotted individuals in Ber- 
muda are the young, the more heavily pigmented, unicolored ones 
being adults. One fact in support of this idea is that small individuals, 
under 100 mm. in length, and living on the lower surface of rock- 
fragments, have but little pigment. In life they are translucent, light 
buff' with only the low, rounded, conspicuous papillae capped abruptly 
with black-brown. It is easy to assume that with increasing size and 
maturity, the deposits of pigment would become heavier and heavier 
until the animal was fully pigmented. In certain individuals however 
the unicolored condition is never attained — just why, it is hard to 
even guess, but large specimens of xanthomela are by no means very 
rare. The largest specimen of badionotus I have ever measured, and 
I believe it is the largest I have ever seen, was a true xanthomela, 
orange-buff, heavily mottled and spotted with black. It was taken 
April 21, 1939, while dredging in only 2 or 3 fms. of water in Ferry 



CLARK: ECHINODERM FAUNA OF BERMUDA 387 

Reach, and measured, in the Aquarium some hours later, more than 
500 mm. long, 110-120 mm. wide and 75-80 mm. high. But complete 
pigmentation may be attained apparently early in life, for "black" 
individuals may be found less than 200 mm. long. The whole matter 
of pigmentation in echinoderms needs and deserves careful and ex- 
haustive study, for it is at the present time a virtually untouched field. 
Research by either the physiological chemist or the biologist or both 
will certainly yield far-reaching and important results. 

8. Thyone surinamensis Semper 

This species was first recorded from Bermuda by Heilprin in 1888 as 
Scmpcria hermudensis sp. but ten years later, I suggested this name was 
a synonym of Cucumaria punctata Ludwig, and in 1901, I stated that 
the species was "not abundant" in Bermuda. In 1907, Verrill called it 
"the most common reef species" of holothurian. In 1926, Dr. Deich- 
mann reported (on the strength of material from Barbados) that Lud- 
wig's species was identical with Semper's Thyone surinamensis of 
1868, and in 1930 she reported it as "abundant at Bermuda", a state- 
ment which I modified in 1933 to "very common". In 1939, however, 
I failed to find a single specimen in more than two weeks of diligent 
collecting. It is ob\'ious therefore that it is not now "abundant", 
except possibly in certain areas which I did not visit. 

9. Thyonepsolus braziliensis (Theel) 

The discovery of a Psolus-like holothurian on the under side of rock- 
fragments at Hungry Bay, April 22, 1937, was the most interesting 
incident of the month's Echinoderm-collecting. The two specimens 
taken measured 13 X 6 X 3 mm. and 23 X 12 X 6. They were not 
found together, but each was living closely appressed to the lower sur- 
face of a rock-fragment, sufficiently large not to be disturbed by ordin- 
ary currents or waves. The rocks were so near low-water mark, it is 
probable they were never quite out of water. The animals were deep 
rosy red (with a purplish tinge), slightly variegated with light gray, 
while the lower surface was light yellowish with but very little pig- 
ment. Considerable pigment has dissolved in the alcohol in which the 
animals are preserved but the deep rosy color is still very obvious. 
The resemblance in coloration to an encrusting bryozoan which 
occurred abundantly in small, roughly circular patches, less than 25 
mm. across, on the lower surface of the rocks near low tide level, was 



388 bulletin: museum of comparative zoology 

so exact that only touching the holothurian could assure the collector 
that it was not a patch of bryozoa. On Monday, April 24, further 
search at the same place resulted in the taking of six more specimens, 
which ranged in size from 10 X 5 X 3 mm. to 18 X 8 X 4. Preserva- 
tion in alcohol results in a shrinkage of 20% or more in length and 
width, but alters the height very little. The coloration is alike in all 
the specimens, and has changed but little in alcohol; the larger speci- 
mens have the rose-color deeper and more tinged with purple than in 
the small specimens. 

Comparison of this material with that in the M. C. Z. from Brazil 
and Tobago, which is labelled Th. hraziliensis, leaves some room for 
doubt as to the identity of the Bermudan specimens. For the older 
material has lost all trace of rose-color, if it were ever present, and is 
simply dirty whitish. There is no hint as to what the color in life of the 
Brazilian form may have been, but the Tobagoan specimens, which 
are all much smaller than those from Bermuda, were collected by me 
in April, 1916, and are recorded in my note-book as "old rose" in color. 
Doubt as to the identity of the Bermudan form is not due so much to 
its notable coloration as to the fact that it has so thick a skin over the 
dorsal scales as to conceal them quite completely while in the Brazilian 
and Tobagoan specimens the scales are very distinct, with only a thin 
skin over them. It is quite possible, therefore, that sufficient material 
will demonstrate that the Bermudan Thyonepsolus is a distinct species. 
Its generic position is clearly indicated by the numerous appendages 
all over the dorsal surface. 

10. Synaptula hydriformis (Lesueur) 

The occurrence of this synaptid in Bermuda in two quite different 
forms has long been known. Verrill considered these forms as distinct 
species. The larger, more striking green and white form he called 
Synapta viridis, while the more oi^ less striped or variegated brown 
form l.e called Sipiapta vivipara. There is no evidence that he ever 
critically compared the two and without such a comparison it is hard 
to believe they are identical. The green and white form is found among 
green zoanthids and algae on the margins of rock fragments or on the 
open sea-bottom. When among zoanthids, it may be relatively short 
and stout, with disproportionately long tentacles (35-50 mm. long, 
5 or 6 mm. in diameter, the tentacles 10 mm. long) while among algae 
it is more slender and the tentacles are shorter. In both cases, the green 
and white (or pale gray) variegation is certainly notably concealing, 



CLARK: ECHINODERM FAUNA OF BERMUDA 389 

SO far as human eyes are concerned. The brown and Hght gray or 
whitish (variegated or more rarely striped) form Hves among brown 
algae and similar growths among rocks or on mangrove roots (or 
branches which are under water at their tips). Both forms are vivi- 
parous and agree in all structural details. 

Dr. Wheeler called my attention to the fact that the dull-colored 
form occurred in a small land-locked body of water on St. George 
Island known as Lovers Lake. This seemed incredible, but he went to 
the lake and brought back specimens, and a visit by Dr. Moore and 
myself subsequently yielded many individuals, of all sizes, from the 
aquatic growth on the mangroves with which the lake is surrounded. 
While there can be no doubt that this lake has subterranean connec- 
tion with the sea, the water is distinctly less saline than that of the 
ocean. The seawater around Bermuda has a salinity of 36.4 while that 
of the lake, kindly analyzed by Dr. E. F. Thompson, was only a trifle 
over 35. The only tangible evidence that this lessened salinity has 
affected the synaptids is that in the larger specimens the anchor plates 
in the bodywall show a tendency to deformation and many are asym- 
metrical. In some specimens, many, perhaps most, of the plates are 
abnormal but other equally mature specimens have the plates very 
generally normal. Apparently the difference from the individuals liv- 
ing in normal sea-water has not yet become a distinguishing mark of 
the Lovers Lake specimens. As the latter are now a practically iso- 
lated colony it will be interesting to note whether, with the passage of 
time, they develop any distinguishing morphological features. 

Both the green and the brown Bermudan Synaptulas differ in general 
appearance in life from the Jamaican form as much as they do from 
each other, but I cannot find any tangible characters by which the 
three forms can be distinguished from each other when killed and pre- 
served. 

IL EuPATiNAPTA ACANTHiA (H. L. Clark) 

This interesting and apparently endemic species was again found in 
April, 1939, in the cove on Coney Island where it was originally taken 
but it was not secured at any other place. There seems to be little 
ground for doubt that the tiny synaptid, only 16 mm. long, which I 
identified in 1899 as inhaerens, was merely a very young example of 
acanthia. At that size, it is probable that two species are not certainly 
distinguishable. 

In spite of the building of the railroad, and other local changes in 



390 bulletin: museum of comparative zoology 

the immediate vicinity, the Httle cove on the northwestern side of 
Coney Island is very much as it was forty years ago, and it was a 
pleasant experience to again collect this fine synaptid at the exact spot 
where I dug it in 1899. Probably it will be found elsewhere, at least in 
Bermuda, under similar ecological conditions, but it is rather odd that 
no other definite locality is known. Heding (1928) records 40 speci- 
mens taken by Mortensen, July 14, 1926, but gives the locality merely 
as "Bermuda: sandy shore at low water". It would be interesting to 
know where this particular sandy shore is. 

12. Epitomapta roseola (Verrill) 

This species was surprisingly uncommon in April, 1939, and all of 
the few specimens taken were very small. Hence I am unable to throw 
any further light on its distribution and habits, or on Heding's variety 
alba, of which Mortensen secured 27 specimens in 1926. Further study 
of this synaptid both at Bermuda and elsewhere is much to be desired. 

13. Chiridota rotifera (Pourtales) 

Apparently as common and widely distributed as it was forty years 
ago, this little holothurian was found plentifully at Achilles Bay, 
Gravelly Bay, Long Bird Island and Hungry Bay. We did not find it 
at the cove on Coney Island, which seems a little odd. 



SUPPLEMENTARY NOTE 

In addition to the above 48 species, it may be convenient to mention 
here the 5 additional species which are known to occur on the Chal- 
lenger Bank, only about ten miles southwest of Bermuda, in water 
some 30 fms. deep. It is noteworthy that 4 of the 5 species are sea- 
stars. They are as follows : 

Chaetaster nodosus Perrier 

Aside from its occurrence at Challenger Bank, where it seems to be 
fairly common, as both the "Challenger" (1873) and the "Gladisfen" 
(1903) secured a number of specimens, this sea-star is known only from 
Guadeloupe, depth unknown, 1 specimen, and from off Havana, Cuba, 
140-200 fms. 



CLARK: ECHINODERM FAUNA OF BERMUDA 391 



Ophidiaster SCHISMOCHILUS H. L. C. 

The holotype alone is known of this notable sea-star. It was taken 
by the "Gladisfen", August 1, 1903, in 30.5 fms. on Challenger Bank. 

Stephanasterias gracilis (Perrier) 

The occurrence of this West Indian sea-star on Challenger Bank in 
about 30 fms. is most interesting, for it is otherwise known only from 
numerous West Indian stations in 56-270 fms. The "Gladisfen" took 
4 specimens in 1903. 

Coronaster briareus (Verrill) 

The "Gladisfen" took a single specimen of this multirayed sea-star 
but there are no data with it. It is a deep-water species of the western 
Atlantic. 

Stylocidaris affinis (Philippi) 

This sea-urchin is common in the Mediterranean and in the West 
Indies in moderately deep water. Its occurrence at Challenger Bank, 
where the "Gladisfen" took one specimen, is not strange therefore, 
and it will not be surprising if more dredging around Bermuda itself 
reveals it as a resident there. 



EXPLANATION OF PLATE 



Clark — The Echinoderni Fiiuna of Bermuda 



Oreaster reticulatus var. bermudensis 
Fig. 1, Upper surface 
Fig. 2. Lower surface 
About one-fourth natural size 



BULL. MUS. COMP. ZOOL. 



Clark: Echinoderm Fauna of Bermuda. 




Bulletin of the Museum of Comparative Zoology 

AT HARVARD COLLEGE 
Vol. LXXXIX, No. 9 



NOTES ON THE UNITED STATES SPECIES 
OF TACHYTES 

(Hymenoptera: Larridae) 



By Nathan Banks 



With Two Plates 



CAMBRIDGE, MASS., U. S. A. 

PRINTED FOR THE MUSEUM 

April, 1942 



(APR 22 1942 / 



L / a K A R y. 



No. 9. — Notes on the United States Species of Tachytes 
(Hymenoptera: Larridae) 

By Nathan Banks 

In 1892 Fox published a revision of this genus and later, 1893, 
added one new species. His division of the males according to the 
structure of the front coxae and front femora is most useful, but the 
separation of the females as "bee-like" or not, is not so helpful. Since 
then Rohwer has described about ten species, Viereck and Mickel 
three, Bradley and I one each. For some years I have tried, at inter- 
vals, to prepare better tables for the females, and to include some of 
those later described, and some which appear to be new. 

Patton, whose descriptions are extremely good, noted that mandi- 
hularis had more spines on the front basitarsi than some other forms. 
This has been used abroad in synoptic tables. However the extra spine 
is sometimes shorter or absent, but is helpful at times. 

On the upper edge of the hind and usually mid basitarsi are usually 
two spines before the tip of the joint, in several species there are three 
of these spines, and they are usually broader, and somewhat curved; 
I have utilized these, and also made a table for those having this char- 
acter, although they are somewhat variable. 

Fox noted that in aurulentus the legs were bare; many species have 
the femora hairy on sides and beneath; however, there are a number of 
species in which the hind femora have no (or only a few) erect hairs 
beneath. This appears to be a fairly natural character. Rohwer 
(Ent. News, 20, 197, 1909) arranged the females in three sections 
according to the color of the scales on the pygidium. The bronzy (often 
partly black) ones are readily distinguished; those with golden or whit- 
ish scales are not so readily separated, for some of those with golden in 
one view may appear partly whitish in another view, and some con- 
sidered white may show some yellowish. However, the grouping is 
helpful and I have made tables to each group. 

The males will doubtless be more surely distinguished by a study 
of the male genitaha. I have figured some of these. I have given 
tables to one section of the large males, and to the small forms, in order 
to include species described since Fox. The shape of the last ventrite 
(8th) in the male is often very useful in separating closely allied species; 
in the small forms it shows less difference. The distance apart of the 
lateral points of this ventrite, and the amount of excision varies some- 
what, and when these points are lobes, as in calcaratus, the breadth 
of the lobe may vary a little. 



396 bulletin: museum of comparative zoology 

The comparative length of the third and fourth antennal joints, al- 
though fairly constant, is often about the same in allied species. Like- 
wise the comparison of vertex-width with the length of second plus 
third antennal joints is sometimes not satisfactory, for when the third 
joint is longer the vertex may be broader. 

The approach of the recurrent veins toward each other, noted in 
species with a partly red abdomen is also variable; likewise the width 
of the second and third submarginal cells on the radius is also variable. 
In some species the third submarginal cell is very long, but it is more 
of a group character. In some of the small species there is a triangular, 
usually polished, depression above the ocellar region, but it is not 
definite enough in all cases to be useful. I have noted that the hind 
basitarsus in some species has one or more spines on the anterior 
(outer) side, while in others there is none except at the tip, and in these 
forms the basitarsus is slightly concave on outer side; this character 
may prove to be more important with further studies. 

In those with hairy hind femora and in some others the groove on 
the posterior slope of the propodeum is widened near middle and above; 
in most of the small forms it is not or scarcely widened, but in some of 
these {parvus, jiattoni) it is widened in the male. In some the groove 
plainly reaches the end of the propodeum. 

Turner has used the length of the galea compared with length of the 
scape. In most museum specimens the galea is not extended, but in a 
number that I have seen there does not appear to be much variation 
in our species, many are shorter than the scape, a few as long or a little 
longer. A more useful structure is the tongue (or labium) and the size 
and extent of its lobes; I have figured these for several species. 

The sculpture in closely related species is but little different, and 
not easily described; the second ventral in our species is dull and 
densely and finely punctate, except on the hind border. Here there is 
a curved row of long bristles, usually four on each side, at least one 
(crassiis) showing five or six. In one (aurulcnfus) the fourth from 
center is more removed from the others than usual. Various other 
minor structures could, doubtless, be studied with profit. 

I have examined the types in Philadelphia and Washington, and 
the M. C. Z. possesses paratypes of some of Cresson's and Rohwer's 
species. Several changes have been made in names as will be ex- 
plained under each species. 

I have proposed a new genus for Tachytcs mergus and an allied new 
species. The species which have a distinct line or groove on the dorsum 
of the propodeum are evidently closely related, and could form a sub- 



banks: tachytes 397 

genus, if the character were constant; some specimens of T. harpax 
and T. columbiae show it, in others faint or absent. 

I ha\ e placed Tachytes mer(,us Fox and a related new species in a 
new genus, Tachyoides, which can be distinguished as follows: 

Basil joint of antennae of female very long, fusiform, more than 
one-half o^ basal width of face; pygidium with but few scales, 
so that the ground color shows between the scales; male 
similar to Tachynana, except somewhat narrower vertex 
(c5. p. 409) Tachyoides 

Basal joint of antennae shorter in both sexes, not fusiform, 
not one-half 1 asal width of face; pygidium covered with 
scales Tachytes 

The type of Tachytes is T. tricolor Panzer (T. europaea Kohl). 
The hind femora are bare, two spines on the hind basitarsus above, 
base of abdomen rufous, and it is very similar to the species I have 
described as T. cressoni. Dahlbom made a genus Tachyptera for 
T. obsoleta Rossi and other European species. This is a black-bodied 
species, four silvery bands, two spines on hind basitarsus, hairy hind 
femora, and a faint groove above on propodeum and thus is very 
similar to our T. mandihularis. The name Tachyptera is preoccupied, 
and since those with the hairy hind femora are easily separated I 
propose a new subgeneric name (Tachyplena) for this section, and 
another (Tachynana) for the small species. 

I desire to thank Dr. F. E. Lutz and Dr. H. Schwarz of the American 
Museum for loaning material; Prof. C. T. Brues for specimens from 
his private collection, and Dr. C. S. Brimley for the opportunity to 
examine specimens from his collection. 

The best grouping of the females that I can suggest now is the 
following: 

1. Hind femora plainly with pendent hairs along most of its lower 
edge; median groove on posterior slope of propodeum broad- 
ened above; but two spines above on upper edge of hind 
basitarsus before tip, and with one or more spines near middle 
of outer (anterior) side; pleura and sternum with long hair. 
Subgenus Tachyplena subgen. no v. genotype T. mandihul- 
aris Patton. Other species included are validus, praedator, 
belfragei, calcaratus, calcaratiformis, harpax, columbiae, crassus, 
distinctus, badius, foxi, auricovianes, ermineif^, floridanus, and 



398 bulletin: museum of comparative zoology 

comanchc; the last six may form a separate group; they have 
four spines on front basitarsus (others five), the tibiae are 
black (in others rufous), and the hind basitarsus sometimes 
has three spines above. 

In all species the males have no groove on base of front 
femora, and no projection at tip of front coxae. 

la. Hind femora practically without pendent hair beneath, except 
possibly a little near base 2 

2. Median groove on posterior slope of propodeum not or scarcely 
widened near middle or above; hind basitarsus with one or 
two spines near middle of anterior (or outer) side, and with 
one or two spines above before tip, rarely three; third joint of 
antennae often no longer, frequently shorter than the fourth 
joint; hind tibiae with the spines above usually short and 
often stubby; mostly small species. Subgenus Tachynana 
subgenus nov, type Tachytes ohscurm Cresson; includes also 
abdominalis, ohducfus, hirkmanni, pattoni, maestus, arizonicus, 
parvus, atomus, and several species known only from males, 
7ninor, minutus, intermedms, hirsutifrons, and amiculus, and 
probably avstcrus. 

2a. Median groove on posterior slope of propodeum widened near 
middle and above; third joint of antennae as long or longer 
than fourth, and usually twice as long as second; spines of 
hind tibiae above sharp, not stubby; frequently three spines 
above on hind basitarsus before tip, and in several species no 
spines near middle of front side of basitarsus; large species. 
Subgenus Tachytes, type, the European T. tricolor; contains 
aurulcntus, austrinus, exornatus, ritfofa.sciatus, elongatus, 
cressoni, fulvivcntris, utahcnsis, hesperm, brevipilis, pepticus, 
sayi, and sericatus. 



Females of subgenus Tachyplena 

1. Tibiae rufous; five spines on front basitarsus, two on hind basi- 

tarsus; median fine on propodeum above usually distinct. . . .2 

Tibiae black; four spines on front basitarsus; often three on hind 

basitarsus 12 

2. But three silvery bands on abdomen; the pygidium bronzy. . . .3 
With four silvery bands on abdomen; wings yellowish 4 



banks: tachytes 399 

3. Bands on abdonnen often faint except on sides; mesonotum 

bordered with gray pubescence; face rather dull golden; fourth 

and usually third tarsal joints without spines crassus 

Bands on abdomen distinct; mesonotum bordered with golden, 
face bright golden; third and usually fourth tarsal joints with 
spines distinctus 

4. Pygidium bronzy to black; face golden; thorax with more or less 

golden pubescence mandihularis 

Pygidium golden to whitish 5 

5. Pygidium whitish or silvery 6 

Pygidium golden to yellowish 7 

6. Silvery bands bend forward near middle of each segment; face 

more silvery ; southern helfragei 

Silvery bands follow margin of segments; northern 

calcaratiformis 

7. Propodeum densely clothed with golden pubescence, also on 

pleura, and sides of mesonotum, and fully as bright on face; 
femora hardly more than one-half black; bands on abdomen 

often somewhat yellowish praedator 

Not so much golden on propodeum, and if face somewhat golden 
the femora black to near tip 8 

8. Face silvery white seen from in front; third joint of antennae 

plainly longer than fourth 9 

Face more or less golden to yellowish 10 

9. Pygidium golden calcaratus 

Pygidium whitish calcaratiformis 

10. Body broad, pygidium broad, bright golden; third antennal 

joint plainly longer than fourth; last joint of antennae tapering 

and longer than penultimate validus 

Body and pygidium more slender; pygidium not so bright golden; 
third joint of antennae but little if any longer than the fourth 1 1 

11. Hind tibia with a distinct dark streak on inner side; clypeal 

margin with a distinct median narrow somewhat bilobed 
projection; third joint of antennae a little longer than fourth 

harpax 

Hind tibia without distinct dark streak; clypeal margin with a 

broad but low projection; third joint of antennae scarcely 

longer than fourth columbiae 



400 bulletin: museum of comparative zoology 

12. Pygidium coppery to black; body with but little silvery pubes- 

cence; third antennal joint hardly longer than the fourth 

hadius 
Pygidium whitish or golden 13 

13. Pygidium whitish to yellowish; body and legs somewhat silvery 

pubescent; third antennal joint hardly longer than fourth; 

last tarsal joint pale foxi 

Pygidium yellowish to golden; body and legs strongly silvery; 
third antennal joint longer than fourth; last tarsal joint dark 

er milieus 

Females of subgenus Tachytes 

1. Wings very dark; but two silvery bands on abdomen; pygidium 

bronzy to black aurulentus 

Wings hyaline to yellowish, at least three silvery bands on 
abdomen 2 

2. But three silvery bands on abdomen; pygidium bronzy to black; 

rarely spines between rows on upper side of hind tibiae 3 

Four bands on abdomen 4 

3. Femora largely rufous; face somewhat golden elongatus 

Femora largely black; face more whitish; bands on abdomen 

broad austrinibs 

4. Tibiae and much of femora rufous; abdomen mostly rufous, large 

species; pygidium golden exornatus 

Femora mostly black, tibiae also 5 

5. Abdomen wholly black; pygidium bronzy to black 6 

Abdomen rufous, or yellowish, at least on basal part above .... 8 

6. Pleura and propodeum strongly silvery sericeous, hair on pro- 

podeum above very short sericatus 

Pleura and propodeum only slightly if at all silvery sericeous; 
hair on propodeum longer 7 

7. From the Middle and Eastern States pepticus 

From Kansas and Nebraska west sayi 

8. Pygidium golden to yellowish or paler; basal segment of abdomen 

at most with only short hair 9 

Pygidium bronzy to black, basal segment plainly hairy at least 
on anterior sides 12 



banks: tachytes 401 

9. Two segments of abdomen rufous ; hair on propodeum moderately 
long; usually but two spines above on hind basi tarsus before 

tip; pygidium pale yellowish cressoni 

Three segments or more rufous 10 

10. Outer side of hind basi tarsus without a spine near middle; no hair 

on basal segment; body rather slender; abdomen usually wholly 

rufous or yellowish rufofasciatus 

Outer side of hind basi tarsus with one or two spines near middle; 
some faint or sparse, fine hair on basal segment; abdomen 
broader 11 

11. Hair on propodeum no longer than tibial spines; propodeum less 

sloping behind, from Texas brevipilis 

Hair on propodeum about twice as long as tibial spines; propod- 
eum steeply sloping behind; from Washington and Oregon 

hesperus 

12. Pleura and sides of propodeum with much silvery pubescence; 

bands on abdomen distinct fulviventris 

Pleura and sides of propodeum scarcely at all silvery; bands on 
abdomen not noticeable except toward sides utahensis 



Females of subgenus Tachynana 

1. Abdomen wholly black, only fine bristles or none on posterior 

side of front tibiae 2 

Abdomen more or less rufous or yellowish; distinct though short 
spines on posterior side of front tibiae 5 

2. With but three silvery bands on abdomen; pygidium bronzy to 

black; spines above on hind tibiae very short and stubby 

ohscuriLs 

More or less white on four segments, though not always plainly 

in a band 3 

3. Pygidium bronzy or coppery; abdomen shining, very small 

.parims 
Pygidium golden, larger species 4 

4. Abdomen shining, with silvery marks mostly on sides, hardly 

bands maestus 

Abdomen dull, with somewhat yellowish bands, almost complete 

obductus 



402 bulletin: museum of comparative zoology 

5. Bands or spots of silvery on three segments only 6 

Bands or silvery spots on the fourth segment also 7 

6. Face silvery white; bands on abdomen white, thorax marked with 

silvery white; pygidium bright golden arizoniciis 

Face dull golden or yellowish; bands on abdomen slightly yellow- 
ish; thorax and pleura with gray pubescence; pygidium rather 
pale golden, sometimes whitish toward base pattoni 

7. Spines on front (outer) side of hind basitarsus; hair on anterior 

sides of basal segment of abdomen; wings somewhat yellowish; 
hair on sides of mesonotum yellowish; spines above on hind 

tibiae very stubby abdominalis 

No spines on front of hind basitarsus; no hair or very faint on 
basal segment; wings hyaline; hair on sides of mesonotum 
white 8 

8. Antennae short; third and joints beyond not twice as long as 

wide at tip; pygidium coppery; six or seven millimeters long 

atomus 

Antennae longer; third and joints beyond fully twice as long as 

wide at tip; pygidium golden to almost coppery; ten to twelve 

millimeters long birkmanni 



Artificial keys to females 
Females with pygidium coppery or bronzy to black 

1. With but two segments with silvery bands, femora and tibiae 

black; all femora bare; wings dark aurulentus 

With three segments with apical silvery band 2 

With four segments with apical silvery band 6 

2. Tibiae black; thorax without white pubescence, no hair under 

hind femora; tibial spines short and stubby; pygidium with 

erect hair besides the depressed iridescent ones obscurus 

Tibiae rufous; no erect hair on pygidium 3 

3. Femora yellowish or rufous; wings yellowish elongatus 

Femora black 4 

4. No erect hair under hind femora; three spines on hind basitarsus 

austrinus 

Hind femora hairy beneath; but two spines on hind basitarsus 

above before tip 5 



banks: tachttes 403 

5. Wings plainly yellowish; pronotum with yellowish pubescence; 

two whitish spots on mesonotum distmctus 

Wings scarcely yellowish; pronotum with white pubescence; no 
spots on mesonotum crassu^ 

6. Hind femora plainly hairy beneath 7 

Hind femora bare 8 

7. Tibiae and tarsi largely yellowish; a median groove on propodeum 

above mandibularis 

Tibiae and tarsi largely black; no median groove on propodeum 
above badiu.s 

8. Less than ten millimeters long; hind basitarsus with but one 

spine above; pygidium coppery 12 

More than ten millimeters long; hind basitarsus with three (or 
two) spines above; pygidium bronzy to black. 9 

9. Abdomen rufous on base 10 

Abdomen wholly black 11 

10. Thorax and propodeum with much white pubescence . .fulviventris 
Thorax and propodeum with little if any white pubescence 

utahensis 

11. Entire body more or less silvery pubescent, or with a sericeous 

bloom in certain views ; hair on propodeum very short . sericatus 
Very little silvery pubescence; hair on propodeum much longer 

pepticus group 

12. Abdomen reddish on basal part atomus 

Abdomen wholly black parvus 



Females with a golden pygidium 

Abdomen partly red ; practically no hair under hind femora, and 
other femora have little but appressed scale-like hair below; 
dorsal surface of propodeum without a median groove 2 

Abdomen wholly black; often long hair on under side of all 
femora 9 

Femora almost wholly rufous; mesonotum bordered with yellow 
pubescence; hind basitarsus usually with three spines above 
or golden exornatus 

Femora (and often tibiae also) almost wholly dark 3 



404 bulletin: museum of comparative zoology 

3. Hind basi tarsus with three spines above before tip; third joint of 

antennae plainly longer than fourth 4 

Hind basi tarsus usually with but two spines above before tip; 
third antennal joint scarcely, if any, longer than fourth 5 

4. Hair on propodeum scarcely, if any, longer than the spines on 

upper side of hind tibiae brevipilis 

Hair on propodeum mostly twice as long or more than the spines 
on upper side of hind tibiae hesperus 

5. Fourth abdominal segment black and without silvery or pale 

band; wings hyaline 6 

Fourth segment with at least some trace of a silvery or pale 
band 7 

6. Face and clypeus silvery white arizonicus 

Face and clypeus yellowish to golden pattoni 

7. Wings hyaline; band on abdomen slightly yellowish, face with 

yellowish to golden pubescence; mesonotum with silvery ap- 
pressed hair on disc; sides of pygidium slightly convex 

hirkvianni 
Wings more or less yellowish 8 

8. Pygidium triangular, sides straight; face white; tibiae often 

partly rufous rufofasciatus 

Pygidium more slender; last two segments of abdomen black; 
tibiae and most of tarsi black abdominalis 

9. Hind femora without hair below; tibiae black; rather small 

species 10 

Hind femora hairy below ; tibiae usually rufous (except ermineus) 

11 

10. Wings somewhat yellowish; abdomen above finely pilose, not 

shining obdudus 

Wings not yellowish; abdomen above largely shining, and white 
bands scarcely noticeable maestus 

11. Tibiae black (with silvery sheen); body very largely silvery 

ermineus 
Tibiae rufous or yellowish see no. 7 of Tachyplena p. 406 



banks: tachytes 405 

Females with a whitish or silvery pygidium 

1. Tibiae pale; long spur of hind tibiae longer than basi tarsus, at 

least a little 2 

Tibiae black; long spur of hind tibiae not longer than basitarsus . 3 

2. Propodeum with more yellowish pubescence; second, third, and 

fourth silvery bands bending forward toward middle, away 

from hind margin of the segment belfragei 

Propodeum with more grayish pubescence, all bands following 
the hind margin of segment calcaratiformis 

3. Abdomen reddish on first two segments above, although not 

wholly so; gray hair on propodeum, hair on basal abdominal 

segment very short cressotii 

Abdomen wholly black; white hair on propodeum; hair on basal 
segment of abdomen moderately long foxi 

Females with three spines on hind 
basitarsus, and hind femora bare 

1. Abdomen with silvery bands on but two segments; wings very 

dark; pygidium bronzy aurulentus 

Abdomen with silvery bands on only three segments; pygidium 

bronzy 2 

Abdomen with silvery bands on four segments, or the abdomen 

partly rufous 4 

2. Legs, including tibiae and most of tarsi, black; some specimens 

of obscurus 

Legs with tibiae rufous or pale 3 

3. Femora largely yellowish or rufous elongatus 

Femora black, except tip aiistrinus 

4. Femora yellowish or rufous; abdomen partly rufous. . .exornatus 
Femora black, and usually tibiae 5 

5. Abdomen at least partly rufous 6 

Abdomen black 10 

6. Pygidium bronzy 7 

Pygidium golden 8 

7. Silvery pubescence on pleura fulviventris 

Little if any silvery pubescence utahensis 



406 bulletin: museum of comparative zoology 

8. Third antennal joint scarcely longer than fourth; abdomen almost 

wholly yellowish, some specimens of rufofasciatus 

Third antennal joint plainly longer than fourth; abdomen with 
apical half black 9 

9. Hair on propodeum above extremely short brevipilis 

Hair on propodeum moderately long hcsperus 

10. Thorax with much silvery pubescence; hair above on propodeum 

very short sericatus 

Thorax almost devoid of silvery pubescence; hair above on pro- 
podeum moderately long 11 

11. From Kansas and Nebraska west sayi 

From Illinois east and south pepticus 

Males of subgenus Tachyplena 

1 . Tibiae largely black 11 

Tibiae mostly rufous 2 

2. Hind and mid tarsi without spines or only minute ones 3 

Hind and mid tarsi with normal spines 5 

3. Wings plainly yellowish; hair and pubescence of thorax plainly 

golden distinctus 

Wings not so plainly yellowish; hair more white or gray 4 

4. Eighth ventrite ends in widely separated points; sixth ventrite 

without short erect hair each side mandihularis 

Eighth ventrite with points closer together; sixth ventrite with 
patch of short, erect, black hair each side crassus 

5. Only three segments bordered with silvery; face bright golden 

auricomanes 
Fourth segment with silvery 6 

6. Some segments near tip with a tuft of hair each side on venter, 

sometimes a trace of a fifth band on abdomen 7 

No tufts of hair on venter near tip, nor any silvery on fifth 
segment 8 

7. With two tufts each side (sometimes rubbed) ; last ventrite ends 

with a very narrow notch in middle, late summer harpax 

With but one tuft each side; eighth ventrite has a broad triangu- 
lar excision between the large, triangular teeth; early summer 

columbiae 



banks: tachytes 407 

8. Next to last ventrite with an apical, transverse band of dense, 

erect, dark hair; face whitish; third joint of antennae longer 
than fourth; last ventrite ends in a broad triangular median 

excision validus 

No such band of hair, or face golden 9 

9. Face golden; third antennal joint scarcely longer than fourth; 

wings yellowish praedator 

Face whitish; third joint of antennae plainly rounded below ... 10 

10. Third joint of antennae hardly longer than fourth; no distinct 

median process to clypeus; eighth ventrite ends in broad, blunt 

tooth each side; southern calcaratus 

Third antennal joint longer than fourth; clypeal margin with a 
small, median, bilobed process more or less distinct; eighth 
ventrite with the teeth plainly more pointed; northern 

calcaratiformis 

11. Hind femora hairy below to tip; parameres slender tapering to a 

fine tip 12 

Hind femora hardly hairy beyond middle, parameres broad, 
flattened 13 

12. Mesonotum not bordered with pale pubescence floridamis 

Mesonotum bordered with pale pubescence ermineu^ 

13. Parameres broad at tip; third joint of antennae hardly twice as 

long as the second comanche 

Parameres pointed at tip; third joint of antennae plainly more 
than twice as long as the second badius 



Males of subgenus Tachytes 
(joints of antennae not rounded below) 

1. Front femora with a groove across base on under side; front 

coxae with projection at tip 2 

Front femora without such groove; front coxae without pro- 
jection 7 

2. Tibiae yellowish or rufous 3 

Tibiae black 5 

3. Abdomen above with some reddish in bands rufofasciatus 

Abdomen black, except silvery bands 4 



408 bulletin: museum of comparative zoology 

4. Four silvery bands on abdomen ; hind femora partly black 

elongatus 

Three silvery bands on abdomen; hind femora nearly wholly 

rufous Seminole 

5. But three bands on abdomen; wings very dark on costal part 

aurulentus 
Four silvery bands on abdomen 6 

6. Third joint of antennae longer than fourth; front side of hind 

basi tarsus without spines near middle; wings yellowish . . apache 

Third joint of antennae scarcely equal to fourth; a spine on front 

side of hind basi tarsus near middle; wings hyaline. . .sericatus 

7. Abdomen more or less reddish on basal part 8 

Abdomen wholly black, except the silvery bands 13 

8. Eleventh joint of antennae widened on inner side, next joints 

more slender fuhiventris 

Eleventh joint not widened at tip 9 

9. Twelfth and thirteenth joint plainly broadened, thirteenth 

hardly twice as long as broad at base spatulatus 

These joints not noticeably broadened, the last tapering 10 

10. Tibiae and hind femora reddish exornatus 

Tibiae and hind femora black 11 

11. But two segments reddish hasirufus 

Three segments reddish 12 

12. Thorax and tip of abdomen deep black utahensis 

Thorax and tip of abdomen grayish black hesperus 

13. Eleventh joint of antennae widened at tip; Eastern species 

pepticus 
Eleventh joint not widened at tip; Western species sayi 

Males of subgenus Tachynana 

1. Tibiae and tarsi yellowish 2 

Tibiae black, tarsi of tep partly dark 3 

2. Face with no erect hair, only short fine pile; base of abdomen 

reddish above; groove of posterior slope of propodeum widened 
above pattoni? 



banks: tachytes 409 

Face with erect white hair; no reddish on abdomen; groove of 
posterior slope of propodeum not noticeably widened above 

minor 

3. Hair on vertex about as long as scape of antennae .... hirsutifrons 
Hair on vertex shorter 4 

4. Abdomen with silvery bands on but three segments; groove on 

posterior slope of propodeum widened above; hair in ocellar 

region white parvus 

Abdomen with some white or yellowish on sides of fourth seg- 
ment 5 

5. Abdomen with some white or yellowish hair on sides of fifth 

segment also 6 

No white or yellowish hair on sides of fifth segment 7 

6. Abdomen shining; hair on vertex moderately long maestus 

Abdomen dull ; hair on vertex very short ohductus 

7. Antennae very short, fourth and fifth segments and most of these 

beyond not twice as long as wide at tip; very small, 5 to 7 

millimeters 8 

Fourth and fifth and some other joints of antennae fully twice as 
long as wide at tip 9 

8. Groove on posterior slope of propodeum widened V-shape above 

TTiinutus 

Groove on posterior slope of propodeum not noticeably widened 

above amiculus 

9. Vertex width hardly more if any than one-third the basal width 

of face Tachyoides mergus 

Vertex width much more than one-third basal width of face ... 10 

10. Face with silvery hair above antennae quite long; tarsi mostly 
dark; a bunch of short brown hair each side on ventral seg- 
ments near tip of abdomen obscurus 

Face with dense silvery appressed hair, little erect above an- 
tennae; tarsi beyond basi tarsus usually pale; abdomen more 
shining above, shorter hair on basal segment intermedius 



410 bulletin: misfaim of comparative zoology 

Genus TACHYTES 

Subgenus TACH\TLENA 

Tachytes iviANDiBULARis Patton 

Common in the Eastern States, mostly north of Carohna, but one 
specimen from Florida. Specimens from Nebraska, Colorado, and 
Texas have more ^.ellow on the femora and are usually larger, and 
represent T. jjrojnnquus Rohwer; I saw the type in the National Mu- 
seum. Males from Nebraska externally appear the same as eastern 
specimens, but preparations of the internal genitalia may show differ- 
ences. The type (National Museum) of T. duplicatus Rohwer from 
Florida seems to be mandibularis. 

Tachytes validus Cresson 

It was described from Texas, the female with golden pubescence on 
the face, and the apical segment (pygidium) bright golden. Fox says 
face with bright silvery pile and pygidium clothed with silvery pile, 
but in the Belfrage collection in the National Museum are specimens 
marked as type which agree with the Cresson description; these, I 
believe, are the true T. validus. It would appear that the specimens at 
Philadelphia are not types; and that there has been some mixup. 

Tachytes breviventris Cresson 

In the M. C. Z. collection are two male Tachytes purchased by 
Packard from Belfrage (along with other Fossorial Hymenoptera). 
The smaller one has on pin "420", and a long folded label "420. Tachy- 
tes breviventris Cress, n. sp.". The other has on pin "418", and on the 
long folded label "418. Tachytes validus Cress, n. sp.". The smaller 
has more rufous on femora, otherwise they are alike; the eighth ven trite 
ends in two lobes, rather near together; on the sixth ventrite a low 
brush of dark brown erect hair, each side on fifth ventrite a small 
patch of similar hair, but hardly as high. The clypeus is the same in 
both, so I consider T. breviventris a synonym of T. validus. 

Tachytes calcaratus Fox 

Described from Florida and New Jersey; I have seen the types in the 
National Museum. I have examined specimens from Miami, De Fun- 



banks: tachytes 411 

iak Springs, and Monticello, Florida, all taken in October, also from 
Raleigh, in September, Sanford and Aberdeen, all North Carolina, 
and the last two taken in October. 

It is very close to T. calcaratiformis, but the female with a distinctly 
golden pygidium, and the male lacks the bilobed process on clypeal 
margin, and the third antennal joint scarcely longer than the fourth. 

Tachytes calcaratiformis Rohwer 

Described from Van Cortlandt Park, New York City; I have not 
seen the female type, but Rohwer says it is like T. calcaratus except 
the pygidium is more silvery. This form is common in southern New 
York; specimens seen from Bronx Park and Bronx ville, New York 
(^ity, Sandy Hook, Long Island, Fisher's Island, Gardiner's Island, 
Mosholu, Nyack, and New Baltimore, New York, also from Province- 
town, Cambridge, Needham, Mass., and Narragansett, Rhode Island. 
Externally the male appears to be the same as T. calcaratus, having the 
same spine-like process on each side of base of the eighth ventrite and 
this ventrite ending in lobes with convex outer sides. 

Tachytes calcaratiformis var. coloradensis var. nov. 

In color and markings similar to the typical form, the pygidium is 
white, the clypeal projection is a little broader and not as plainly 
bilobed; the pygidium has a longer slender apical part. 

In the male the third antennal joint is hardly a bit longer than the 
fourth (in calcaratiformis plainly longer) and as much rounded below 
as the fourth; the tip of the eighth ventrite is narrower than in the 
typical form, and the lobe each side is more pointed. 

Length 9 13 mm., d' 11 to 12.5 mm. 

Holotype, Boulder, Colorado, August (Carpenter), M.C.Z. no. 
25629; paratypes from White Rock, near Boulder, 13 August, and 
Wray, 17 to 19 August, both Colorado, in Amer. Mus. Nat. Hist, and 
M.C.Z. 

Tachytes belfragei spec. nov. 

This, I believe, is what Fox in his Revision calls the female of T. 
validus, with silvery pile on face and pygidixim. 

It is near to the northern calcaratiformis and to calcaratus (except 
in silvery pygidium). Face with silvery white hair, thorax with white 
hair and pubescence; abdomen with four silvery bands; these bands 



412 bulletin: museum of comparative zoology 

do not follow the hind margin of the segment all across, but near mid- 
dle bend forward; the pygidium white; femora black, rest of legs 
rufous; wings yellowish as in allied forms. The third joint of antennae 
is plainly longer than the fourth; the propodeum has a median groove 
above; hind basitarsus has two spines above, not as long as those of 
calcaratiformis, the front basitarsus has five or six spines; propodeum 
densely long-haired ; the pygidium has straight sides (in calcaratiformis 
a little concave before tip). 

Length 9 17 mm. 

From Texas (Belfrage coll.) (Peabody Acad.). Acquired by Packard 
when he was at the Salem institution. Type M. C. Z. no. 25630. 

Tachytes harpax Patton 

Described from Waterbury, Conn., from both sexes, the female with 
golden pygidium; Wheeler took many at Colebrook, Conn. Fox puts 
as female one with silvery pygidium, probably calcaratiformis. This, 
doubtless, led Rohwer to describe dubitatus, which appears to be a 
female of harpax, but I have not seen the type. 

Other specimens are from Englewood, New Jersey, Falls Church, 
Virginia, Lexington, Holliston, Dorchester, Scituate, and Wellesley, 
Mass.; North Conway, New Hampshire; Milwaukee, Wise; and Mid- 
land County, Michigan; mostly taken in late July, August, and Sep- 
tember. Some males show traces of a white band on the fifth segment. 

Tachites columbiae Fox 

This species is closely related to T. harpax. The hind femora are 
hairy below, hind basitarsus with two spines, front basitarsus with five; 
dorsal groove sometimes indicated, not as plain as in mandibulatus, 
praedator, etc. In most males there is a narrow white band at tip of 
fifth segment, at least on sides. The eighth ventrite has two sharp, 
triangular teeth at tip, rather near together. 

Specimens have been taken at F'alls Church and Glencarlyn, Vir- 
ginia, in June and July; Beltsville, Maryland, 6 July; Da Costa, New 
Jersey, 28 July; Raleigh, 28 May, 8 June, Judson, 19 July, and Edge- 
combe Co., 23 June, all North Carolina. 

Tachytes praedator Fox 

This is one of the most brightly marked species, the golden of head 
and pygidium being very bright; the bands on abdomen are usually 



banks: tachytes 413 

slightly golden, sometimes strongly so; the wings more or less yellowish. 
The legs are largely yellowish, the front and mid femora often not one 
half black, and the hind femora black only near base. The hind femora 
are plainly hairy below, the hind basitarsus with two spines, the front 
with five, and a distinct dorsal groove on the propodeum. In the male 
the antennae are simple, the eighth ventrite rather slender and with 
two broad blunt teeth not far apart; hind tarsi with spines at tips of 
joints, the basitarsus with two or three other spines. It is not uncom- 
mon at Fedor, Texas; also found at Beltsville, Maryland, 6 July; Wil- 
mington and Raleigh, North Carolina, early July; and a male from 
between Climax and Bainbridge, Georgia, 28 July (Amer. Mus. Nat. 
Hist.). 

Tachytes floridanus Rohwer 

I examined the type from Florida in the National Museum. It is 
much like a small T. pepticus, but the hind femora have distinct, 
though fine, white hair below, another male agreeing with this is from 
Southern Pines, North Carolina, 17 May (American Museum Natural 
History), another from Southern Pines, 29 May (N. Car. Dept. Agric), 
one from Aberdeen, North Carolina, 29 May (N. Car. Dept. Agric). 
The apical portion of antennae simple, the third joint a little longer 
than fourth; the hair on propodeum and basal segment of abdomen is 
long and white; the surface of propodeum shining; its posterior slope 
has a very broad triangular depression. I have seen no female to fit it. 

Tachytes ermineus spec. nov. 

Body and legs black; face densely covered with silvery white pub- 
escence and erect hairs; antennae with silvery bloom, third joint plainly 
longer than the fourth, last joint tapering; thorax and propodeum with 
silvery pubescence and white hair; hair on mesonotum as long as that 
on vertex, that on propodeum very dense and longer; abdomen above 
with four fairly broad silvery bands, faintly interrupted in middle, 
basal parts of these segments with a silvery sheen, fifth segment at tip 
narrowly rufous, with black and some rufous hair; pygidium moder- 
ately slender, densely covered with appressed pale golden hairs, the 
dark margin of pygidium very prominent. 

Front and mid legs behind with much appressed and some erect 
white hair, hind femora shining, but with some fine white erect hairs 
below; outer side of hind tibia and three joints of tarsus bright silvery, 
the spines here, as elsewhere, pale yellowish, on hind tibiae not extend- 



414 BITLLETIN: MUSEUM OF COMPARATIVE ZOOLOGY 

ing on basal third; on upper edge of hind basitarsus but two spines 
before tip; front basitarsus with four rather long spines, and those on 
the next two joints still longer and very slender. Wings hyaline, veins 
yellowish, third submarginal cell extending beyond the marginal, the 
recurrent veins not especially close above. Clypeus broadly rounded 
below, no prominent teeth, but the lateral one distinct. 

Male very similar in appearance; third joint of antennae a little 
longer than fourth, last joint tapering; legs with hair as in female, the 
hind femora with distinct fine white hair below; spines on legs more 
whitish; abdomen does not show the silvery sheen on basal part of seg- 
ments, last segment wholly black, pygidium rather more tapering than 
usual; eighth ventrite with two moderately large triangular teeth 
nearly touching at inner base. 

Length female 13 mm., male 11.5 mm. 

From Lower Ranger Station, Pima Co., Catalinas, ^\rizona, 6 to 
20, and 15 to 21 July, about 2700 ft. 

Type and allotype in American Museum Natural History, paratype 
in M. C. Z. 25633. 

T. ermineus bears a very close resemblance to T. sericatus, but is 
separated by the pale golden pygidium (bronzy to coppery in sericatus), 
with but two spines above on hind basitarsus (three in sericatus), and 
by the hind femora hairy below (bare in sericatxis). 

Tachytes crassus Patton 

The hind femora are hairy below; two spines on hind basitarsus 
above, usually five on front basitarsus; a distinct groove on propodeum 
above. In male the hind tarsus has few if any spines; the eighth ven- 
trite ends in two large triangular teeth, and before tip are tufts or 
brushes of short erect black hair (in mandihularis no such tufts, and 
efghth ventrite ends in two more widely separate points). In both 
sexes the abdomen appears black with scarcely noticeable silvery bands, 
these being narrow; in mandibularis the bands are distinct in both 
sexes. 

Specimens seen from Wollaston and Waltham, Mass.; from Wash- 
ington Co., Wisconsin; and Cambridge, Nebraska. 

Tachytes auricomanes Bradley 

The types (two males), which I have examined, are similar, as Brad- 
ley states, to T. crassus, but differ in having but three silvery bands on 



banks: tachytes 415 

abdomen, the fourth segment wholly black, and the hind tarsi with the 
spines at tips of the joints. Face golden; mesonotum not as promin- 
ently bordered with golden as in T. distinctus, propodeum with golden 
pile and long gray to j'ellowish hair; basal segment of abdomen with 
long white hair on basal part; femora black, but extreme tip and rest of 
leg beyond rufous. Hind femora with very distinct pendent hair on 
under side; no spine on outer side of hind basi tarsus, none above; long 
spur of hind tibia not nearly equal to basitarsus. 

Third antennal joint little longer than fourth, the fourth, fifth, sixth, 
and seventh joints a little rounded below. 

On the sixth ventrite there is short erect hair across, not plainly 
higher and denser at each side; the eighth ventrite ends in two large, 
rather sharp pointed, triangular teeth. 

Length 14 to 16 mm. 

From Oglethorpe, Ga., 1 July (J. C. Bradley), in Cornell University 
collection. 

This might be the true T. distinctus Smith, but Smith states that the 
mesonotum is bordered with golden, a character not so obvious here 
as vnth the insects I have interpreted as T. disfindiis. 

Tachytes distinctus Smith 

This is the species, similar to T. elongates, but with black femora, 
which Smith described from Georgia. Fox applied the name to Smith's 
"var. B", which is elongatus; the name must go with the typical form. 
I append a new description. 

9 Body dull black; face with yellowish to golden pubescence; 
antennae, including scape, black; mesonotum margined with yellowish 
pubescence, also each side of scutellum; on front part of mesonotum 
are two elongate, slightly separated silvery or yellowish spots of 
pubescence; pleura with a large spot of yellowish to golden pubes- 
cence; propodeum with rather long gray to yellowish hair, on each side 
an elongate patch of golden pubescence; abdomen with three silvery 
bands, about as broad as in T. elongatus, base of first segment with long 
white hair; pygidium bronzy to coppery. Femora black, except ex- 
treme tip, which with the tibiae and tarsi are fulvous; front and mid 
femora with appressed yellowish hair behind, outer side of hind tibiae 
golden ; wings plainly yellowish with brownish yellow venation. 

Face much narrowed above, vertex hardly wider than width of 
antennal bases; third joint of antennae plainly longer than fourth; 
elypeus with two small teeth each side; scutellum with a faint groove 



416 bulletin: museum of comparative zoology 

behind; propodeum with median groove above; pygidium much as in 
T. elongatus; front basitarsus with four spines before tip, hind basi- 
tarsus with two spines above before tip, but none on outer side, spines 
at tips of tarsal joints of usual length ; inner spur of hind tibia about as 
long as basitarsus; all femora hairy below, the hind femora with rather 
short, fine white hair. 

The male is similar in color, but face more golden, and no submedian 
spots on mesonotum ; four silvery bands, all narrow, and that on fourth 
segment only toward sides and very narrow; pygidium silvery; legs 
with femora more broadly fulvous at tip; the hair on mesonotum very 
short and gray. 

Clypeal margin with three or four small teeth each side; antennae 
with fourth, fifth, sixth, and seventh joints convex below and slightly 
concave above. Mid and hind tarsi almost without spines, tips of one 
or two joints with one or two very short ones, and basitarsus with two 
or three minute ones above. Eighth ventrite ends in two teeth, scarcely 
sharp, rather far apart at tips, close together at base. On the sixth 
ventrite is a tuft of black hair each side, between the tufts the hair is 
lower, and less dense. 

Length female 18.5 mm.; male 16 mm. 

Female from Nome, Liberty Co., Texas, 22 June (Bequaert), male 
from Shierer, Terrebonne Co., Louisiana, 18 June (Bequaert); Raleigh, 
N. C, 12 July, Maysville, N. C, 11 July, and Marion, N. C, 8 July, 
taken by Brimley and Mitchell. 

* 

Tachytes foxi spec. nov. 

Body, legs, antennae dull black, last tarsal joints rufous; abdomen 
above with an apical silvery band on four segments; tibial spurs 
rufous; the spines on tibiae and tarsi white. Wings clear, veins 
fulvous. 

Face and pygidium white; moderately long white hair on thorax, 
propodeum, and basal segment of abdomen; longer white hair on face, 
and short dark hair on vertex. Third joint of antennae plainly longer 
than the fourth; middle of clypeal margin slightly but broadly pro- 
jecting and nearly truncate, no distinct lateral teeth; propodeum above 
without a median groove, the apical pit connected to the broad furrow 
on posterior slope. 

Pygidium moderately broad at base, apex rounded. All femora have 
some white hair below; on hind basitarsus are but two spines before 
the tip; the inner spur of hind tibia as long as the basitarsus. 



banks: tachytes 417 

Length, 9 , 12 mm. 

Two females from Fedor, Lee Co., Texas, 6 May, 1909, and 17 May, 
1898 (G. Birkmann). Type M. C. Z. no. 25634. 

Tachytes comanche spec. nov. 

Male. Black, legs black, except apical tarsal joints; face and base of 
mandibles with white hair; no silvery on pleura; hair on mesonotum 
short and dark, on propodeum long and white. Under hind femora 
there is some hair at least to middle. Four segments of abdomen with 
silvery apical band. Propodeum shows an apparent dorsal groove; 
antennae short, third joint a little longer than fourth. Eighth ventrite 
ends in two triangular points, rather far apart. Parameres are broader 
at tip than in hadius; at the subbasal outer angle there are two long 
stout bristles, and in front for a short distance the edge is slightly con- 
cave; beyond are other, shorter bristles. 

Length 12 mm. 

From Fedor, Lee Co., Texas, 8 and 11 May (Birkmann). Type 
M. C. Z. no. 25694. 

Tachytes badius spec. nov. 

Black body, also antennae and legs except the pale last tarsal joint; 
wings hyaline. Face silvery, lower part of occiput also, mesonotum 
bordered with silvery, also pleura, posterior sides of propodeum^ and 
four bands on abdomen; femora silvery below, and mid and hind 
tibiae above; pygidium coppery; long white hair on upper part of face, 
on propodeum, and on basal segment of abdomen. Clypeus with a 
slight process, rounded below, and somewhat sinuate each side before 
the lateral angle; third antennal joint a little longer than fourth; no 
median groove on propodeum above, behind with a moderately wide 
groove; pygidium moderately broad, with plainly convex sides and 
almost pointed tip; front basitarsus with four spines above before tip, 
hind basitarsus with three spines; spines on hind tibiae pale, slender; 
inner spur of hind tibiae equal to basitarsus; hind femora distinctly 
hairy below, as also other femora. 

Length, 9 , 13 mm. 

One female from Anhalt, Comal Co., Texas, 28 June (Bequaert). 
Type M. C. Z. no. 25632. 

It differs from T. pepticus in hairy hind femora, clypeal margin, 
white border to mesonotum, small pit on propodeum, and in hair on 



418 bulletin: museum of comparative zoology 

basal segment of abdomen about twice as long as in T. pepticus. The 
male, which I believe belongs to this species, since it has hair on under 
side of hind femora and occurs in the same region, is generally similar to 
pepticus, with short dark hair on mesonotum and long white hair on 
the propodeum. The third joint of antennae is a little longer than the 
fourth, the tenth and eleventh not broadened; the hind femora have 
white hair below, not dense, and little, if any, on apical third; the 
eighth ventrite has a large, rather sharp-pointed tooth each side, the 
emargination rather deep. The apical part of the paramere is broad, 
flattened above, angulated near base on outer side, and in front of this 
angulation a row of stiff, curved hairs, the lower margin (seen from 
side) is nearly straight; the penis has the tip very slender (seen from 
above), and not swollen much near middle. 

Length of cf 1 1 to 12 mm. 

From Lee Co., Texas, 17 May; Florence, Arizona, 28 July; Mesilla 
Park, New Mexico, 12 July; Col ton, California; and Low^ell Ranger 
Station. Pima Co., Arizona, 6 to 20 July (A.M.N.H.). 



Subgenus TACHYTES 
Tachytes aurulentus Fabr. 

This large dark-winged species is readily known by having but two 
silvery bands in the female, and but three in the male. The femora 
are bare; the femora and tibiae dark, tarsi more or less rufous; the 
pygidium bronzy to black. The hind basitarsus ( 9 ) has three spines 
above, front basitarsus with five before the apical one; the pit on the 
propodeum extends forward more than in many species, but there is no 
real dorsal groove; the antennae of the male not modified. 

It is found in the Southeastern States, but has been recorded from 
New Jersey, and I have taken it at Falls Church, Virginia; collecting 
dates are usually from the end of July to end of September. 

Tachytes elongatus Cresson 

This species was regarded by Fox as the male of distindus. Smith 
had two forms under distinctus, one, the principal description, with 
black femora, and a "variety B" with yellowish femora. Fox admits 
his distinctus is the variety B. However, the specific name cannot be 
applied to the variety. As localities Smith gave Philadelphia and 
Georgia. Smith's description of black femora agrees with a southern 



banks: tachytes 419 

species. Thus elongatus remains as the name of the common and wide- 
spread species which has been called distinctus, and the name distinctus 
must be applied to the form that is similar to elongatus but has the 
femora black. In the male the paramere has a large fringe of long 
spreading hairs or bristles at outer tip. It is most common in the 
Southern and Southwestern States, Texas to California, in the East 
rarely north of Washington, D. C, but in the West north to Nebraska 
and Colorado. Most of the far western males seen are smaller than 
the eastern ones. 

Tachytes elongatus var. seminole var. nov. 

Males. Similar to male of typical form, except that there are but 
three silvery bands on abdomen, and the hind femora are almost 
wholly rufous, and the amount of rufous on front and mid femora is 
more than in typical form. These markings are like those in the 
female of elongatus, rather than the male; hair on face white (instead 
of dull golden in elongatus). The eighth ventrite ends in two slender 
points far apart, the parameres have heavier bristles at tip than in 
elongatus, and more spreading. 

Length 14 to 14.5 mm. 

From South Miami, 13 August and Larkins, Florida, Oct. (both 
Graenicher). Type M. C. Z. no. 25692. Also one from Dry Tortugas, 
Florida (Thompson), Amer. Mus. Nat. Hist. 

Tachytes elongatus var. apache var. nov. 

Males. The two specimens are smaller than any T. elongatus, 
although size is variable. The femora are entirely dark (pale at ex- 
treme tip in elongatus); the tibiae also dark, except tip of front and 
mid pairs (wholly pale in elongatus), and tarsi partly darkened; hair 
on face and base of mandibles silvery white, and long white hair on 
propodeum. The body is slender as in elongatus, and the bristles on 
parameres similar. 

Length 12 mm. 

From Patagonia, Arizona, 20 August (Bequaert); and Tucson, 
Arizona (Snow). Type M. C. Z. no. 25693. 

Tachytes austrinus nom. nov. 

T. contractus Fox is preoccupied by T. contractiis Walker 1871, so a 
new name is necessary. Besides the type of T. contractus Fox I have 
seen but one female, this from Monti cello, Florida, 16 August. 



420 bulletin: museum of comparative zoology 

The hind femora are bare; three spines above on hind basi tarsus, 
four on front basi tarsus and no groove on dorsum of propodeum. 

Tachytes sericatus Cresson 

Described from Texas. I have females, but no male that appears to 
belong to it. The female has very short hair on propodeum above and 
on basal segment of abdomen. The pygidium is bronzy, but in one 
view could be called "dull silvery." The hind femora are bare, and 
the hair under mid femora is very short; hind basitarsus with three 
spines above, front basitarsus with four. 

My females are from Fedor, Lee Co., Texas, June to August. 

The male, which has been considered as that of T. sericatus, is much 
more slender than the female; face with snow-white hair, little above 
antennae; hair on thorax very short, but only slightly silvery; propod- 
eum also with rather scant short hair on sides, above rather long and 
pale, on lower sides silvery; legs, with tarsi, mostly black, spines 
whitish, one near middle of outer side of hind basitarsus; abdomen 
shining; silvery bands on four segments; very short hair on basal seg- 
ment, but longer on anterior sides. 

Third antennal joint a little shorter than fourth; vertex rather 
narrow; groove on propodeum widened, but not so much as in female. 
Abdomen slender, eighth ventrite ending in two slightly divergent 
slender prongs; long spur of hind tibiae not quite as long as basitarsus. 

From above the parameres are flat, moderately slender, and ob- 
liquely narrower from outer side to tip. Seen from the side, the lac- 
iniae are long, the short apical part apparently separate, with a rounded 
lobe above near base; the lower side of laciniae fringed with long, fine 
hair. 

Length 10 to 1L5 mm. 

From Fedor, Texas, Valentine, Texas, 8 July, and Steins, New 
Mexico, 14 July (Bequaert). 

Tachytes pepticus Say 

Say gives Indiana as locality; in the collection of the American 
Museum of Natural History is a male from Lafayette, Indiana; this 
male agrees with Say's description in having the pubescence of face 
and the spot at base of mandibles somewhat golden, from above only 
is the face white. Prof. Brues has taken various other specimens near 
Chicago and northern Illinois, Graenicher had a pair from Milwaukee, 
Wisconsin, and Dreisbach has taken it in Michigan. 



banks: tachytes 421 

These are all easily separable from western specimens by the male 
genitalia and the marks of head when not discolored ; the points of the 
eighth sternite are more widely separated. In the male the tenth and 
eleventh joints of the antennae are plainly broadened, the tip of 
eleventh being broader than base of the twelfth. 

The specimen (c?") from Lafayette, Indiana, 16 Aug. 1920, about 
550 ft. alt., I make the neotype; it is in the Amer. Mus. Nat. Hist. 

Besides the localities mentioned, I have seen specimens from Woods 
Hole, Massachusetts; Southern Pines and Raleigh, North Carolina; 
Miami, Florida; Auburn, Alabama. 

I consider T. pennsylvanicvs Bks. based on males from Rockville 
and West Fairview, Pa., July and August, as at most a form of pepticus. 
At the time of description I had not seen the true pejjficus, and Fox 
used pepticv^ for the western species here described as sayi. The 
specimens seen, types and some from St. Louis, Mo., one from Ken- 
tucky (Sanborn), are larger, and the eighth ventrite is plainly propor- 
tionally broader than in the typical pepticus from the Central States; 
the male from Woods Hole is somewhat intermediate, and none of the 
eastern males seen have as narrow an eighth ventrite as the typical 
form. Compared with the width of face below the vertex is narrower 
than in the true pepticus. The parameres are broader toward tip seen 
either from above or from side than in typical pepticus, and less sharp- 
pointed, but in pepticus there is some variation, so I leave it as a variety 
or form of pepticus. I have seen no eastern females that would go with 
it. The males from Fedor, Texas differ slightly from typical pepticus. 
(See Fig. 12.) 

Tachytes sayi spec. nov. 

This is the western representative of pejMcus, which Fox treated as 
pepticus. The male is separated from pepticus by having the base of 
mandibles with white pubescence and the hair on face silvery white in 
most views, except directly in front. The hair on propodeum is fully 
as long as in pepticus. The tenth and eleventh joints of antennae are 
not widened, and the twelfth not narrower. The body (phallus) of the 
male genitalia is more slender than in pepticus, the parameres, though 
curving slightly, are not nearly so curved nor so widespread as in 
pepticus; the tip of each paramere is truncate; the penis is moderately 
slender near tip but not so noticeably widened near middle as in 
pepticus; the eighth ventrite ends in a triangular tooth each side, the 
two not so far apart as in pepticus. The bands and amount of thoracic 
pubescence are about as in pepticus. 



422 bulletin: museum of comparative zoology 

Length of d^ 10 to 12 mm. 

The holotype is from Clear Creek, Colo., taken many years ago by 
Oslar. Others from the same locality; Cambridge and Bartley, 
Nebraska; Hot Springs and Buffalo, S. Dakota; and Lee County, 
Texas, where dated all in July; also Provo, Utah, 20 July to 1 August; 
and Jim Creek, near Boulder, Colorado, 3 August (A.M.N.H.). 
Type M. C. Z. no. 25673. 

The females differ little from pepticus, and I see no structural 
difference. Specimens are from Berkeley, Colorado; Elmo, Kansas; 
Canton, S. Dakota; and Spokane, Washington, mostly August. 

Tachytes fulviventris Cresson 

This was described from a female, Colorado. The type has long hair 
on propodeum. The hind femora are bare; the hind basitarsus has 
three spines above, the front four. The abdomen has three segments 
rufous. Specimens before me are from Dragoon Mts., 20 July and 
Bonita, 12 July, both Arizona, taken by Bequaert; Sierra Blanca, 
El Paso Co., 8 July, and Valentine, 8 July, both Texas and by Be- 
quaert; Davis, Calif., 30 June (Bohart); Finney Co., Kansas, August 
(Snow). Also in American Museum from La Junta, Colo., 12 August; 
Wray, Colo., 17 August; Crook, Colo., 24 August; Sabino Basin, 
Ariz., 8-20 July, and Marfa, Tex., 3-6 July. Recorded from Nebr., 
Mont., and Wash. 

Tachytes fulviventris var. inferioris var. nov. 

This form is distinguished by having but two segments of abdomen 
rufous; the hair on propodeum is as long as in the typical form, and 
other characters similar. 

Length 12 to 13.5 mm. 

This is the variety found in Lafayette, Indiana and northern 
Illinois. Type M. C. Z. no. 25695, also paratype in Amer. Mus. Nat. 
Hist., and two in the Birkmann collection without locality, possibly 
Texas. 

Tachytes brevipilis spec. nov. 

In appearance much like T. fulviventris, the first three segments of 
the abdomen being rufous, but the hair on the propodeum is not one 
half as long, and the pygidium is golden not bronzy. The hair on 
vertex is also only about one half as long as that of fulviventris; the 



banks: tachytes 423 

third joint of antennae is a trifle longer than the fourth, not as much 
so as in ftdmventris. The hair on face and clypeus is plainly yellowish, 
not the silvery of fulviventris. There is some silvery pubescence on each 
side of the propodeum toward tip; on the base of the first abdominal 
segment the hair is also short. The legs are black, but three or four 
tarsal joints rufous; the hind femora are not hairy below, and the other 
femora have a silvery pubescence below; there are four spines on the 
front basitarsus before tip, and three on the hind basitarsus as in 
fulviventris; the long spur of hind tibia is equal to the basitarsus. 

Length 15 to 17 mm. 

Three females from Lee Co., Texas, June (Birkmann). Type 
M. C. Z. no. 25636. The male is not yet known. 

Tachytes hesperus spec. nov. 

Body black, first three segments of abdomen rufous; the hair on 
face is grayish; there are some patches of silvery on pleura and pro- 
podeum but not prominent; the pygidium is yellowish to golden. 
The clypeus is rounded below, with a few small teeth each side; the 
third joint of antennae a little longer than the fourth; the hair on 
propodeum is white, almost as long as in fulviventris, there is no 
median groove on dorsum, the pit is crossed by fine striae. There are 
four spines on front basitarsus before tip, three on hind basitarsus 
above as in fulviventris. The wings are hyaline, venation yellow-brown, 
the third submarginal cell extends beyond the marginal ; no hair under 
hind femora. 

The male is similar, but hair on face is white; the third joint of 
antennae is little longer than the fourth, the third joint from tip not 
widened and the apical joint slender and tapering. The eighth ven- 
trite with narrow points and widely separated, more so than in basi- 
rufus. 

From fulviventris the female is separated by the pale pygidium; from 
brevipilis by having much longer hair on propodeum. The male is 
separated from spatulatus and from the male accredited to fuhiventris 
by the unmodified apical segments of the antennae, as well as by the 
shape of the eighth ventrite. 

Length 9 14 to 16 mm., d^ 12 mm. 

Several from Spokane, Washington, 21, 22 July, Yakima City, 
Washington, 2, 3, 4 July, and Umatilla, Oregon, 24 June, all taken by 
Samuel Henshaw in 1882 ; also from Oregon, in the American Museum 
of Natural History. Types in M. C. Z. no. 25637; paratypes there and 
in A.M.N.H. 



424 bulletin: museum of comparative zoology 

Tachytes utahensis spec. nov. 

Black like T. pepiicus, with silvery face and hardly any pale pub- 
escence on pleura, but the first three segments of abdomen are rufous 
above and below. There is more and longer erect hair on each side of 
the face; the hair on propodeum is long like pepticus, and the pygidium 
is bronzy. The sides of pygidium are slightly concave. The propodeum 
above has no median groove, the apical pit is striated across bottom 
and connected to the broad, but tapering groove on the posterior slope. 
The hind femora are not hairy below; there are four spines on the front 
basitarsus and three on the hind basitarsus; the inner spur of hind 
tibia is almost equal to basitarsus. The wings are hyaline, veins 
brown, the marginal cell is more than four times as long as broad, 
longer than in pepticus, the third submarginal reaches beyond the 
marginal. The clypeus is rounded below, with an obtuse tooth each 
side. 

The male is similar, but smaller and more slender, three segments 
rufous; third antennal joint not longer than the fourth, last joint 
tapering; the eighth ven trite shows two sharp points rather widely 
separated. 

Length 9 15 mm., male 12 mm. 

Type female from Watson, Utah, 29 July (Carpenter), M. C. Z. 
no. 25638. Allotype male from Provo, Utah, 29 July to 1 August, 
American Museum of Natural History. Paratypes from Eureka, Utah, 
13 to 21 July, Glenwood Springs, Colorado, 22 to 29 July, and Yellow- 
stone National Park, 4 August, in M. C. Z. and A.M.N.H. 

Tachytes exornatus Fox 

Described from a male from Las Cruces, New Mexico. It is our 
most handsome species. 

The female (allotype) has the body and legs marked as in the male, 
and likewise the wings are yellowish and darker at tip. The pygidium 
is covered with dense, fine, bright golden hair. There are two spines 
on hind basitarsus above, four on front basitarsus; hind femora with- 
out hair below; no median groove on propodeum above, and third joint 
of antennae plainly longer than the fourth. 

Length of female 16.5 to 18.5 mm. 

Allotype from Tempe, Arizona, 2 August (Bequaert), another, same 
locality, 1 August, and one from Sabino Basin, St. Catalina Mts., 
Arizona, 8 to 20 July (American Museum of Natural History). 



banks: tachytes 425 

Tachytes rufofasciatus Cresson 

Described from a male from Texas (Belfrage). Common at Fedor, 
Lee Co., Texas, in May and June. Also from /Vuburn, Lee Co., Ala- 
bama, 9 June; Southern Pines, North CaroUna, 7 June; MacCoUum, 
Couseta Co., Georgia, 8 June; Monticello, Florida, 25 May; and also 
from Davis and Colton, California; the latter rather larger, but appear- 
ing to be the same species. The males of nifofa^ciatus are usually quite 
dark, with paler bands, the female usually has the abdomen entirely 
rufous or yellowish. The wings always show somewhat yellowish, and 
the pygidium broader at base than in allied species. 

Tachytes cressoni spec. nov. 

Black; face and pygidium silvery ; first and most of second abdominal 
segment dull rufous; abdomen above with four silvery bands; legs 
black, last few tarsal joints reddish; spines on tibiae and basitarsi 
white; upper surface of mid and hind tibiae silvery; femora only 
slightly hairy; head with white hair on face, darker on vertex, short 
and brownish on mesonotum; moderately long and gray on propodeum; 
upper part of sides of propodeum with white pubescence, rather faint 
except at tip; pleura and sternum with white hair. Wings scarcely 
yellowish, the veins yellow. ( 'lypeus slightly swollen below, and 
slightly emarginate in the middle; third antennal joint a little longer 
tlian the fourth; propodeum with a median groove above distinct, on 
posterior slope moderately broad, widened at the turn and here striate. 
Pygidium moderately long, sides not concave, not as broad at base as 
T. rufofasciatus. Hind basitarsus with two spines on upper outer row 
before tip; front basitarsus with four before tip; the upper inner side of 
hind tibiae show five spines before tip, all short. 

In fore wing the lower side of third submarginal cell is plainly longer 
than the outer lower side of the second cell; the long spur of hind 
tibia not longer than the basitarsus. 

Length, 9 , 13 mm. 

From Fedor, Lee County, Texas, 1 and 3 May (G. Birkmann), 
also Austin, Texas, 9 May (Brues). Type M. C. Z. no. 25631. 

Tachytes spatulatus Fox 

This is known from the male only. Specimens before me are from 
places in Colorado, North Dakota, L'tah, Wyoming, and several from 



426 bulletin: museum of comparative zoology 

Oak Creek Canon, Arizona (Snow coll.). The eighth ven trite ends in 
two broad, rather rounded lobes. It is possible that this is the male of 
T. fulviventris, instead of T. caelehs Patton, so placed by Fox. 

Tachytes basirufus Rohwer 

I have seen a type in Washington (U.S.N.M.). It has moderately 
long hair on propodeum; all basitarsi dark, with white spines; male 
only known. The eighth ventrite is rather slender, the apex with two 
broad, blunt teeth, not far apart. The hind femora have a few hairs 
on basal part below. It differs from other western males in shape of 
this ventrite.. Specimens before me are from La Junta, Colorado, 12 
August; BluflF, Utah, 7 July; and Mesilla Park, N. Mexico, 12 July. 
The female is unknown. 



Subgenus TACHYNANA 

Tachytes abdominalis Say 

Described from Arkansas. I apply this name to specimens like the 
two here labelled by Cresson from Dallas, Texas (Boll), and have 
marked as neotype one with the Cresson label, and the number 
"166", and that of Hagen's list "355". These are recorded in Cresson's 
Hymenoptera Texana. Others come from Fedor, Lee Co., Texas 
(Birkmann) and Yt. Stockton, Texas, 5 July (Bequaert). In American 
Museum are two, Grand Junction, ( olo., 17 July, and La Junta, Colo., 
12 August. The legs are wholly black and the spines above on hind 
tibiae are very short and stubby; the pygidium is golden to whitish 
near base; two spines above on hind basitarsus and four on front 
basitarsus. In all specimens seen the rufous occupies the whole of the 
first three segments of the abdomen. Say mentions the white lunule on 
each side of the fourth segment. No male is known that appears 
referable to this species. 

Tachytes obscurus Cresson 

A deep black species with little silvery except on face; bands on but 
three segments of abdomen; the femora are bare, hind basitarsus with 
two spines, front with five or six. 

This is a ver\' common species at Fedor, Texas, June to September. 
Also Valentine, 8 July, Waco and Alpine, 30 Sept., Texas. I have taken 



banks: tachytes 427 

it in Washington, D. C. and Falls Church, Virginia, 14 August and 7 
September; also from Willets, 24 August, and Swannanoa, 12 Septem- 
ber, both North Carolina; Brinson, Georgia; Long Key, Miami, 15 
August, 4 September, both Florida; and Tucson, Arizona. Recorded 
from Nebraska and Kansas. 

T. texanus Cress, is considered the male, and I believe, correctly. 
The eighth ventrite ends in a slender spine each side, rather far apart. 
The fifth and sixth ventral segments at sides toward tip have a brush 
of low but erect brown hair. 

Tachytes obductus Fox 

Type came from Tennessee. Both sexes are common at Fedor, 
Texas in June. The wings are darkened, particularly the marginal cell; 
the bands on abdomen usually a little yellowish. The hind femora 
bare, hind basitarsus with two spines, front with five. 

The male (allotype) is more slender than female, the wings less 
darkened, face silvery, thorax and propodeum nearly covered with 
whitish hair; four segments silvery at tip; antennae short, fourth 
joint a little longer than third; the groove on posterior slope of pro- 
podeum very slender; tibiae with few short, white spines; pygidium 
rather slender; eighth ventrite ends in two sharp points. 

Length 9 to 9.5 mm. 

Tachytes birkmanni Rohwer 

Known only from females which appear to be not uncommon near 
Fedor, Lee Co., Texas. A rather slender-bodied species, with hyaline 
v/ings, silvery bands on four segments, faint in middle, but on sides 
extending forward; the median groove is linear; four or five spines on 
front basitarsus before tip, hind basitarsus with one or two spines 
above, no spine near middle of anterior side; pygidium slender, golden; 
tibial spines not as stubby as in T. abdominalis; third antennal joint 
fully twice as long as the second. 

Tachytes minor Rohwer 

Based on a male, much like T. birkmanni of which it may be the 
male. Third joint of antennae hardly twice as long as second, and not 
longer than fourth; vertex -width fully one half basal width of face; 
groove on posterior slope of propodeum linear; no spine on front 



428 BULLETIX: MUSEUM OF COMPARATIVE ZOOLOGY 

(outer) side of hind basi tarsus; eighth ventrite with two short, rather 
well-separated points. 

Besides the type, I have seen two specimens, also from Fedor, 
Texas, in the Birkmann collection. 

Tachytes pattoni spec. nov. 

This species is similar to both T. rufofasciatus and T. hirkmanni. 
Like them, it is wholly black except the basal part of abdomen, which 
is rufous on three segments, in the other species often rufous on all 
segments. Like them, it has two spines above on hind basitarsus, four 
on front basitarsus, no hair under hind femora, no dorsal groove to 
propodeum, the pygidium is golden, and the spines on hind tibia are 
short and stout. It is smaller than most rufofasciatus, but is broader 
than most birkvianni; the pubescence on face and clypeus is dull 
golden like that on hirkmanni (on rufofasciatus it is white). The bands 
on abdomen are not white as in hirkmanni, but not as yellow as those 
of rufofasciatus, but it differs from both in having but three bands; the 
fourth and fifth segments in specimens seen (ten) are wholly dull black 
and no trace of a band on the fourth. The pygidium is less narrowed 
toward tip, and the sides slightly but plainly convex; the third joint 
of antennae is no longer than fourth; the wings are hyaline, venation 
yellowish; the pubescent patches on pleura and propodeum are hardly 
as prominent as in rufofasciatus and paler. 

Length of body 8.5 to 9.5 mm. 

Several specimens from Fedor, Lee Co., Texas (Birkmann), 17 and 21 
June; one from iVIiami, Florida (A. E. Wight), taken in March, this 
last the holotype, M. C". Z. no. 25639. 

A male from Raleigh, North Carolina, 16 June, might be the male 
of this species. The basal segment is partly reddish above, the tibiae 
and tarsi pale as in T. minor. It is not as slender as minor and there is 
no long erect hair on face, but only very short pile, which is certain, 
views from above appears golden. The third joint of antennae only a 
little longer than the second, and shorter than the fourth; long spur 
of hind tibia about equal to hind basitarsus, latter without spines on 
the anterior side; eighth ventrite ends in two points rather far apart; 
the groove on posterior slope of propodeum is much widened above as 
in -parvus and miiiutus. The hair on mesonotum is short and plainly 
yellowish, that on propodeum still shorter q,nd white; the face does not 
show the two rounded elevations seen in parvus; in fore wings the top 
of the third submarginal cell is wider than the top of the second. 

Length 7 mm. 



banks: tachytes 429 

Tachytes arizonicus spec. nov. 

Black, but abdomen largely reddish, the last two or three segments 
black; legs black, tarsi more or less yellowish; face, back of head, 
borders of mesonotum, pleura, sternum, posterior sides of propodeum, 
and bands on three segments of abdomen silvery; mesonotum with 
dark, faintly yellowish pubescence in middle; pygidium golden; 
wings hyaline. Face with erect white hair, rather short; propodeum 
with moderately short gray hair; no hair beneath hind femora, others 
with white appressed pubescence. Clypeus below in a broad, evenly 
rounded lobe, projecting only a little; third joint of antennae hardly 
longer than fourth ; propodeum without median groove above, that on 
posterior slope very narrow; spines above on hind tibiae very short 
and stout, rufous or brown; inner spur of hind tibiae about equal to 
basitarsus; two spines above on hind basi tarsus, five above on front 
basi tarsus; pygidium moderately broad, Sides slightly convex; in fore 
wing the third submarginal cell reaches a little bevond the marginal 
cell. 

Length ? , 9.5 ram. 

From Tempe, Arizona, 31 July and 2 August (Bequaert). Type 
M. C. Z. no. 25640. 

Tachytes maestus Mickel 

Black; tarsal joints more or less pale; pygidium golden; wings plainly 
a little yellowish; face, borders of mesonotum, pleura, sternum, pos- 
terior sides of propodeum, and four bands on abdomen silvery; some 
white also on last segment; the bands on second, third, and fourth 
segments are widened on sides ; ventral segments also with some white 
pubescence, the abdomen generally shining. Hair on upper face rather 
short, long white hair on slope of propodeum and on base of abdomen. 

Clypeal margin with a broad median lobe, with straight sides, and 
truncate across tip; third joint of antennae no longer than fourth; no 
distinct median line on propodeum above, behind the groove is narrow 
and short; pygidium rather broad, sides scarcely convex, except near 
tip; spines above on hind tibiae pale and sloping; inner spur of hind 
tibia about as long as basitarsus, hind basitarsus with two spines above 
before tip, front basitarsus with five spines above before tip. 

In the male the colors are the same; the abdomen rather slender; 
the front coxae with a small apical tooth; hind tibiae with spines; the 
last ventral ends in two fine, widely separated spine-like points. The 



430 bulletin: museum of comparative zoology 

abdomen is distinctly shining; the third antennal joint is as long as the 
fourth; the ventral plate ends in two widely separated slender points; 
the hair on pygidium is wholly depressed. 

Length 9 9.5 mm., cf 8 mm. 

From Tempe, Arizona, 30 July, 1 and 4 August (Bequaert); type 
was from Nebraska. T. intermedius is said to have suberect pile on the 
pygidium, pubescence sparser than in obscurus (here much more 
abundant); and pile on legs more or less golden (not in maestus). 

I have examined a paratype of maestus in the National Museum. 

Tachytes intermedius Viereck 

I have identified as this species several males from Tempe, Arizona, 
taken by Bequaert in August. The hair on head and thorax is short; 
the clypeus and face below ocelli is bright silvery, the tarsi (except 
basitarsi) rufous; the pygidium is rather broad, the eighth ventrite ends 
in two widely separated sharp points, not unlike ohductus, but more 
widely separated. The second and third abdominal segments are 
rather broadly margined with white, the first and fourth less broadly; 
there is no hair under hind femora, but white appressed hair on 
posterior sides of front and mid femora; the long spur of hind tibia is 
not nearly as long as basitarsus. I have not seen the type which was 
described from Douglas Co., Kansas. 

Tachytes austerus Mickel 

Described from Nebraska; I have not seen it, so it is not in the 
tables; the abdomen is partly red above, and is said to be similar to 
T. ahdominalis, but with the clypeal margin strongly emarginate, and 
the recurrent veins not so proximate above. 

Tachytes hirsutifrons spec. nov. 

cf Dull black, last tarsal joint scarcely paler, abdomen above some- 
what shining; face and clypeus with silvery pubescence; vertex, thorax 
above, pleura, and base of abdomen densely clothed with very long 
grayish white hair, longer than in other species of the size, the hair on 
vertex equal to second plus third antennal joints; abdomen with four 
silvery bands and the pygidium silvery as usual; venter with a tri- 
angular silvery patch on each side of three segments. Front and mid 
femora very hairy below and behind, hind femora not; mid and hind 
tibiae silvery above. 



banks: tachytes 431 

Third antennal joint much shorter than the fourth, last joint 
slender, longer than the preceding; the second plus third, plus fourth 
scarcely, if any, longer than vertex-width. Clypeal margin with a 
moderately broad projection, the margin scarcely rounded; pro- 
podeum with no evident median groove; hind tibiae with two rows of 
very short pale spines above; hind basitarsus with one spine on outer 
side at middle and a minute one above; the inner spur of hind tibiae 
not as long as basitarsus. 

In fore wings the marginal cell is fairly long so that the part of hind 
margin beyond the third submarginal is almost as long as the part 
before this point; the second submarginal cell is much broader above 
than the third submarginal. 

The ventral plate ends in two small, rather widely separated points; 
the last three ventral segments with short, erect black hair; the 
pygidium is rather narrow at tip. 

Length 8.7 mm. 

From Tempe, Arizona, 2 August (Bequaert). Type M. C. Z. no. 
25635. 

About the size of T. intermedius, but with broader body, and dark 
tarsi, the long hair on vertex separates it, as well as from T. texana 
(the male of obscurus). 

Tachytes parvus Fox 

This was described from a male from Camden Co., New Jersey. 
I took a male at Beltsville, Maryland, 6 July; the American Museum 
of Natural History has a male from Valley of Black Mts., 27 August 
(Beutenmiiller), and Dr. C. S. Brimley has taken both sexes at 
Raleigh, North Carolina, cf 29 July, 9 1 July, 13 September, and 
Tarboro, 29 July. The female is similar to the male, scarcely larger, 
black, the lower face and clypeus with white pubescence, and the 
abdomen with white pubescence on the sides of four segments, not 
forming bands; the abdomen is fully as shining as the male; the tarsi 
become rufous toward tip; the pygidium is moderately slender, the 
sides straight, and covered with rather coarse coppery hairs, no erect 
hair. 

The clypeus is rounded below, a faint median excision and two or 
three teeth laterally. In front of the anterior ocellus is a raised area 
each side as in the male; the third antennal joint no longer than 
fourth, and about one half longer than second. The mesonotum has 
some short, dark hair, the propodeum with short, white hair above 



432 bulletin: museum of comparative zoology 

and behind, much shorter than in male; some faint patches of white 
pubescence on pleura and sides of propodeum. 

Front basitarsus with four short spines, hind basitarsus with one or 
two above before tip, and one on the outer side; the hind tibiae have 
moderately long sharp-pointed spines above; the long spur of hind 
tibia fully as long as basitarsus; hind femora have no long hair below 
(as in all related forms). 

Length 6.5 mm. 

The male is easily recognized by having but three silvery bands on 
abdomen, as well as by the raised areas on the face, and its small size 
and darkened wings. 

Allotype 9 from Raleigh, N. Carolina, 1 July 1932 (C. S. Brimley). 

Tachytes minutus Rohwer 

Described from Lee Co., Texas. I have seen a male from Raleigh, 
North C^arolina which agrees with description; it has the groove on 
posterior slope of propodeum widened above as in T. parvus; there are 
four silvery bands, or rather spots, each side on abdomen, and the 
antennae are very short as in amiculus, there is more white hair on 
front femora than in amiculus, and the long spur of hind tibia is equal 
to the basitarsus (shorter in amiculus) ; the eighth ventrite has longer 
processes at the tip than in amiculus; I have seen the type in Wash- 
ington. 

Tachytes amiculus spec. nov. 

cf Black, last two or three tarsal joints yellowish; wings hyaline; 
face with short appressed silvery hair, front and vertex with longer 
erect hair, fine and pale; thorax with scattered white hairs, but no 
distinct borders, also some white on mesopleura and sternum; pro- 
podeum with short white hair on sides, in middle the hair is gray and 
indistinct. Abdomen above shining, basal segment without erect hair; 
white bands, broadly interrupted, on four segments, extending for- 
ward on sides; pygidium silvery. 

Vertex almost one half basal width of face, clypeal margin rounded 
in middle, with a small tooth each side; antennae very short, second 
joint about two thirds of third, beyond the joints are not twice as long 
as broad at tip. Propodeum above deeply and closely punctate right 
up to the tip, where, instead of the spear-shaped fovea, appears a 
faint, angulate ridge; posterior slope also deeply punctate, the median 



banks: tachytes 433 

groove linear, not widened above in a V-shape. Femora more slender 
than in T. parvus, no hair on hind femora beneath ; inner spur of hind 
tibia almost equal to basitarsus, latter without a spine on anterior 
side. 

Eighth ventrite with a rather short, blunt tooth each side, tipped 
with a hair, not as slender nor as sharply pointed teeth as in T. minutus. 
Top of second submarginal cell longer than top of third. 

Length 6 to 6.5 mm. 

Type from Cold Spring Hfirbor, Long Island, N. Y., 24 July, on 
beach; paratypes from Raleigh, N. Carolina, 25 June, and Edgecombe 
Co., N. Carohna, 23 June in North Carolina Dept. Agriculture col- 
lection. 

Differs from description of T. minutus in lacking the V-shaped 
widened groove of posterior slope of propodeum, in no impunctate 
area at end of upper side of propodeum ; in more blunt processes at tip 
of eighth ventrite. 



•&' 



Tachytes atomus spec. nov. 

A minute species; black, first three segments of abdomen rufous, 
tips of tarsi pale, pygidium coppery. 

Lower face and clypeus with short, silvery hair, hair on upper part 
of face and on vertex extremely short, silvery back of eyes, also the 
usual elongate silvery band each side on collar. Mesonotum with 
silvery hair on sides bj- wings and behind, in middle with short bronzy 
hair; pleura as well as femora and tibiae silvery; abdomen with silvery 
spots on sides; propodeum behind silvery on the sides, sparse short 
hair above. Antennae with third joint no longer than the fourth; 
propodeum above with a very faint scar at tip, as in other small species 
the middle tip seems to project in a short point, groove on posterior 
slope very narrow. Abdomen slender, with sparse white hair below and 
toward tip above; hair on propodeum extremely short; but one spine 
above on hind basitarsus before tip, hind femora bare. Wings hyaline; 
marginal cell only a little more than three times as long as broad; 
shorter than in T. hirkmanni; the third submarginal cell does not ex- 
tend beyond the second further than the second beyond the end of 
third discoidal cell; in one specimen the two recurrents are very close 
together above. 

Length 9 6 to 7 mm. 

From Tempe, Arizona, 31 July (Bequaert). Type M. C. Z. no. 
25641. 



434 bulletin: museum of comparative zoology 

Tachyoides gen. no v. 

The female of Tachytes mergus Fox (obscuranus Rohwer) differs 
from normal Tachytes in the long fusiform basal joint of the antennae; 
the clypeal margin has a much larger tooth each side, and the pygidium 
has but few scale-like hairs so that the dark ground color of the pygi- 
dium shows between the scales; the hind femora have no hair below, 
and that on other femora is short. The bands on abdomen are very 
broad on the sides. The groove on the slope of propodeum is linear, 
scarcely widened above. In the male the vertex is very narrow. 

Besides the genotype, 2'. mergus, the genus includes one new species. 

The male having much resemblance to Tachynana is separated in 
the table to males of that subgenus (cf. p. 409). The females of the 
t\\ o species separate as below : 

Antennae with second joint about half the length of third, latter 
fully equal to fourth; basal joint longer than second, third and 
fourth joints together ariella 

Antennae with second joint not nearly one-half of third, latter 
longer than fourth; basal joint not longer than second, third, 
and fourth joints together mergus 

Tachyoides mergus Fox 

Tachytes obscuranus Rohwer is a synonym, as is noted in the Wash- 
ington collection. The female was described from Camden Co., New 
Jersey, and Rohwer described both sexes of obscuranus from Fedor, 
Lee Co., Texas, where it appears to be fairly common in June and early 
July, the males sometimes in late May. 

The female has four spines on front basitarsus before tip, the hind 
basitarsus two spines above before tip, and one on outer side near 
middle. There is usually white scale-like hair on basal part of the 
second abdominal segment (not seen in Tachytes). 

The male has no spines above or on outer side of hind basitarsus; 
the vertex is only about one third of the basal width of face; the eighth 
ventrite ends in two rather long, well separated sharp points. 

Tachyoides ariella spec. nov. 

Black, tips of tibiae and the tarsi yellowish; wings hyaline, veins 
yellowish. Face brilliantly silvery, back of head also; mesonotum with 



banks: tachytes 435 

anterior corners and a faint subniedian stripe each side silvery; pleura, 
mid and hind femora beneath, tibiae on outer side, propodeum with a 
stripe each side, more or less plainly silvery; four abdominal segments 
with a large, silvery triangle each side, scarcely meeting in middle, and 
extending beneath. Pygidium with silvery scales, shorter than in 
T. mergus. Hind femora bare, other femora with some short erect hair 
beneath; spines on legs slender and pale; front basi tarsus with four 
spines above, hind basitarsus with two; spines on hind tibiae above 
slender, inner spur of hind tibia fully as long as basitarsus. Clypeal 
margin with a stout, large tooth each side, much as in T. mergus. 
Antennae short and slender, basal joint longer than in T. mergus, 
longer than the second, third, and fourth joints together; second joint 
about one half of the third, latter as long as the fourth. Propodeum 
above shining, with a distinct median groove, pit not very broad, and 
on the posterior slope the groove is rather narrow. Pygidium tri- 
angular, sides straight. 

Length 7 mm. 

Female from Tempe, Arizona, 1 August (Bequaert). Type M. C. Z. 
no. 25642. 



436 bulletin: museum of comparative zoology 

BIBLIOGRAPHY 

. Banks, N. New Nearctic Fossorial Hymenoptera. Ann. Entom. Soc. 
Amer., 14, p. 18. 1921. 

Bradley, J. C. A new Tachytes from Georgia. Ent. News, 30, p. 298, 1919. 

Ckesson, E. T. Catalogue of Hymenoptera in the Collection of the Entomo- 
logical Society of Philadelphia from Colorado Territory. Proc. Ent. Soc. 
Phila., 4, 1865, Larridae, pp. 464-467. 

Cresson, E. T. Hymenoptera Texana. Trans. Amer. Ent. Soc, 4, Larridae, 
pp. 213-218, 1872. 

Fox, W. J. Monograph of the North .\merican species of Tachytes. Trans. 
Amer. Ent. Soc, 19, pp. 234-252, 1892. 

Fox, W. J. The North American Larridae. Proc. Acad. Nat. Sci. Phila., 
1893, pp. 467-551. 

MiCKEL, C. E. New species of Hj^menoptera of the subfamily Sphecoidea. 
Trans. Amer. Ent. Soc, 42, pp. 399-434, 1916. 

MicKEL, C. E. A .synopsis of the Sphecoidea of Nebraska. Univ. Nebr. 
Studies, 17, no. 4, pp. 342(24)-456(138), 1917. 

Patton, W. H. List of a collection of Aculeate Hymenoptera made by 
Mr. S. W. Williston in Northwestern Kansas. Bull. U. S. Geol. Geogr. 
Survey, 5, no. 3, pp. 349-370, 1879. 

Patton, W. H. List of North American Larradae. Proc. Bost. Soc. Nat. 
Hist., 20, pp. 385-397, 1880. 

RoHWER, S. A. Notes and descriptions of wasps of genus Tachytes. Ent. 
News, 20, pp. 197-206, 1909. 

RoHWER, S. A. New Hymenoptera from Western United States. Trans. 
Amer. Ent. Soc, 35, pp. 99-136, 1909. 

RoHWER, S. A. Sphecoidea, in Guide to Insects of Connecticut, Part III, 
Hymenoptera, pp. 645-697, 1916. 

RoHWER, S. A. Descriptions of six new wasps (Hym.). Proc. Ent. Soc. 
Wash., 22, pp. 54-59, 1920. 

Say, Thomas. Leconte Edition, 1, p. 165; 2, p. 754. 

Smith, Frederick Catalogue of Hymenoptera [of British Museum], 4, 
Fossores, p. 307, 1856. 

Viereck, H. L. Notes and descriptions of Hymenoptera from the Western 
United States. Trans. Amer. Ent. Soc, 32, p. 211, 1906. 

Williams, F. X. Monograph of the Larridae of Kansas. Kans. LTniv. Sci. 
Bull., 8, no. 4, pp. 121-213, 1914. 



EXPLANATION OF PLATES 



PLATE 1 



Banks — ^Tachytes 



PLATE 1 

Fig. I. Tachytes pepticus, psLTamere, above. 

2. Tachytes seminole, tip of paramere and brush. 

3. Tachytes apache, tip of paramere and brush. 

4. Tachytes elongatus, tip of paramere and brush. 

5. Tachytes obscuriis, paramere, above. 

6. Tachytes sayi, paramere, above and from side. 

7. Tachytes badius, paramere, above. 

8. Tachytes sericatus, paramere, above. 

9. Tachytes sericatus, paramere, side. 

10. Tachytes badius, paramere, side. 

11. Tachytes comanche, paramere, above. 

12. Tachytes pepticus, paramere of Fedor specimen above and from side. 

13. Tachytes ermineus, paramere, above. 

14. Tachytes pennsylvanicus, lacina from side. 

15. Tachytes pepticus, paramere Miami specimen. 

16. Tachytes pennsylvanicus, paramere, from side. 



BULL. MUS. COMP. ZOOL. 



Banks. Tachytes. Plate 1 






PLATE 2 



Banks— Tachyti's 



PLATE 2 

17. Tachytes pepticus, apical ventrite of neotype. 

18. Tachytes pepticxis, tip of abdomen from above, Michigan specimen. 

19. Tachytes pepticus, apical ventrite, Miami specimen. 

20. Tachytes pennsylvanicus, apical ventrite. 

21. Tachytes crassus, apical ventrite. 

22. Tachytes sericatus, apical ventrite. 

23. Tachytes Jloridanus, apica\ veninte. 

24. Tachytes amiculus, tip of abdomen, above. 

25. Tachytes auricomanes, apical ventrite. 

26. Tachytes elongatus, apical ventrite. 

27. Tachytes ermineus, apical ventrite. 

28. Tachytes, badius, apical ventrite. 

29. Tachytes sayi, apical ventrite from below and above. 

30. Tachytes sayi, apical part of antenna. 

31. Tachytes pepticus, apical part of antenna, neotype. 

32. Tachytes pennsylvanicus, apical part of antenna. 

33. Tachytes fulviventris, tongue, side and above. 

34. Tachytes praedator, tongue, side and above. 

35. Tachytes obscurus, apical ventrite, two specimens, above. 

36. Tachytes rufofasciatus, tongue, above and from side. 

37. Tachytes amiculus, part of antenna. 

38. Tachytes obscurus, apical ventrite from below. 

39. Tachytes floridanus, hind femur. 

40. Tachytes aurulentus, tongue from above. 

41. Tachytes mandibularis, tongue, from side and above. 



BULL. MUS. COMP. ZOOL. 



Banks. Tachytes. Plate 2 




H.CT. -L 



Bulletin of the Museum of Comparatiye Zobloffy 

AT HARVARD COLLEGE 

Vol. LXXXIX, No. 10 



REVISION OF THE AFRO-ORIENTAL GECKOS 
OF THE GENUS PHELSUMA 



The Library 

MusQua of Comparative Zoology 

Harvard University 

By Arthur Loveridge 



CAMBRIDGE, MASS., U. S. A. 

PRINTED FOR THE MUSEUM 

June, 1942 




'UUM IS 1942 



No. 10. — Revision of the Afro-Oriental Geckos of the Genus Phelsuma 

By Arthur Loveridge 

Not since Boulenger (1885d, p. 209) monographed this genus fifty- 
six years ago has any revision been attempted. At that time 8 species 
were recognized, the same number as Mocquard (1909, p. 19) included 
in his key to the Malagasy species a quarter-of-a-century later. 
Rendahl (1939a, p. 263), in connection with a report on a collection 
from the Seychelles, gave a resume of 13 species and described 6 more, 
of which unfortunately I can recognize but two. Dr. Kendall was 
severely handicapped by the very scanty comparative material at his 
disposal. 

In this respect the Museum of Comparative Zoology, with 17 of the 
23 species now recognized, is more fortunately placed. This is largely 
due to the energetic and far-sighted policy of Dr. Thomas Barbour who 
assembled them over a period of years. It is with particular appropri- 
ateness and pleasure, therefore, that I associate his name with the only 
novelty here described: 

Phelsuma barbouri spec. nov. 

A paratype of this new species is in the American Museum of Natural 
History, whose entire Phelsuma collection was generously placed at 
my disposal by Mr. C. M. Bogert. I have also to thank Dr. D. Cochran 
for submitting a species in the United States National Museum, and 
Mr. V. FitzSimons of the Transvaal Museum for answering questions 
regarding the type of standingi. Under present conditions it is not sur- 
prising that queries regarding certain species addressed to European 
museums have gone unanswered for the letters may not have reached 
their destinations. 

One conclusion that must remain questionable is my action in re- 
ferring cepediana to the synonymy of inunguis (vide pp. 7, 11). For 
many years half-a-dozen species were assigned to cepediana (1820) by 
the earlier herpetologists and resulted in much confusion, a confusion 
largely due to the homogeneous character of the lizards comprising 
this difficult genus. 

In this connection it might be as well to point out the relative im- 
portance of the characters here employed for separation of species. 

While snout length does assist in defining certain groups, its value 
has been, and is, largely masked by the vagueness with which it was 
formerly expressed. Its usefulness will be fully established only when 



440 bulletin: museum of comparative zoology 

long series of age and sex of each species have been recorded. The 
presence, or absence, of a median cleft on the rostral is extraordinarily 
constant in several species of which good series are available; in 
others, however, it appears to be of little value. Apparently only in 
the Mauritius group of islands does one find the nostril bordered by 
the rostral and the centre of the nostril immediately above the suture 
between rostral and first labial. So far only two species (mutabilis and 
breviceps) have been recorded with less than the standard number of 3 
nasals. While in a majority of species the normal number of inter- 
supranasal granules is 2, with 1 or 3 as less frequent variants, in 
several well-known species a single granule appears constant. The 
number of labials, whether upper or lower, appears of little conse- 
quence excepting for two species (mutabilis and breviceps) where reduc- 
tion is correlated with a shortening of the snout. 

Boulenger (1885d) sometimes designated the dorsals "granular 
scales;" as I have been unable to know where to draw the line between 
those species where such a term is reasonably applicable and those 
species where "granules" is more appropriate, I have used the shorter 
term throughout. It is interesting to note that the ventral scales are 
smooth in all species except trilineata, lineata, bimaculata, quadrio- 
cellata and astriafa, and in some of these so obtusely as to be scarcely 
noticeable at times. The number of subdigital lamellae of the fourth 
toe serves only to differentiate the two species mutabilis and breviceps. 
The subuniform or transversely dilated character of the subcaudals 
appears to me to offer the most important indication of relationships 
which we have, so that it is unfortunate that it was not employed by 
Rendahl when describing sundbergi and other forms from the Sey- 
chelles. 

As a result of these studies several species have had to be referred 
to the synonymy, viz. 

Gecko cepedianus Merrem = ? P. inunguis (Cuvier) 

Phelsumia lineata var. bifasciata Boettger = P. lineata Gray 

Phelsuma carinatum, Rendahl = P. astriata Tornier 

Phelsuma carinatum mahecnse Rendahl = P. astriata Tornier 

Phelsumia laticauda var. comorensis Boettger becomes P. dubia como- 

rensis Boettger 
Phelsuma pulchrum Rendahl = P. m. longinsulae Rendahl 

Phelsuma pulchrum cousinense Rendahl = P. w. longinsulae Rendahl 
Phelsumia micropholis Boettger = P. mutabilis (Grandidier) 

In Madagascar there appears to be a fairly well-marked division as 
between those species found in the drier west and those occupying the 



loveridge: afro-oriental geckos 



441 



MT. D'AMBRE 



NOSY BE 



NOSY SAKATIA ,<^ o 



MAKAMBY ID. o 





COL PIERRE RADAMA 
ANDRANOLAVA 



MAJUNGA 
KANDRANY (KANDANI) '"^ STE. MARIE de MAROVOAY 
' SOALALA (SOALA) 
^0° • MEVATANANA FAN DRARAZANA 




LAKE ALAOTRA 
ANTSIHANAKA f) 

rVOLINA 
DIDY SWAMPS 



AMBATONDRAZONAKA 



SAKANA 
STE. MARIE ID. 
FENERTVE 
ILE AUX PRUNES 
TAMATAVE 
TAMPINA 
AMBILO 



ANAMALOZOTRA 



BETAMPONA RESERVE 



FIANAR ANTSOA 



ANOVOKA 



ANDRINGITRA MTN3. 
, ANJAMALA (UNZAMALY) .-;>' 

TSrVONOINA V . ANDRANOHINALY 

TULEAR V ANDRANOLAHO 
MAROAMALONA J* TONGOBORY • MANASOA 
STE. AUGUSTIN 




AMBILA 



LAKE 



• AMBINANY (EMINIMINY) 



MANAMPETSA (TSlMANAMPETSO)i 



ANDRANOVAHO 



AMPANIHY 
, ANDROY 
V ANKARIMBOLA 



FORT DAUPHIN 



Map of Madagascar showing approximate position of all localities 
in which Phelsuma have been collected 



442 BULLETIN MUSEUM OF COMPARATIVE ZOOLOGY 

mountainous and more humid east, though towards the extreme north 
and south such a division tends to disappear. In view of the difficulties 
I have encountered in locating many Malagasy places, largely owing 
to the haphazard way in which prefixes and suffixes are attached to 
root names, a map is furnished showing the approximate positions of 
localities as located on the best map known to me, that of G. Grandidier 
and J. Hanson (1925, Pub. by Soc. d'Edit. Geog. Marit, et Coloniales). 
The spellings on that map are accepted and those of herpetologists 
placed in parentheses, except for the following: 

Andranohinaly (Boettger), though probably ANDRANOHINALAHY 
(of the map). Andranovaho (Angel) though quite possibly ANDRANO- 
MAVO (of the map). *Angurutani (Boettger) of Northwest Mada- 
gascar, which I failed to find. Ankarimbola (Ankarimbela : Boettger) 
given as KARIMBOLA (on the map). Col Pierre Radama (Mertens) 
not found unless COL AMBATOND RADAMA (of map). *Fierin 
(Grandidier, 1869) though probably FIRINGA, S. W. Mad. (of map). 
*Tanosy (Kaudern) not found, though obviously near Manasoa and 
Tulear. Tsianovoka (Angel; Anevoka: Boettger) not found though 
S.S.W. of Fort Carnot and Mt. Ikongo in eastern Madagascar. 

Genus Phelsuma' 

1825. Phelsuma Gray, Ann. Philos., 26, p. 199 (type Gecko cepedianus 

Merrem). 
1830. Anoplopus Wagler, Nat. Syst. Amphib., p. 142 (type Gecko cepedianus 

Merrem). 

Digits clawless or with a vestigial claw, not webbed, thumb and 
inner toe vestigial and not dilated, remaining digits subcylindrical at 
base, below covered by scales which distally are replaced by trans- 
versely dilated shields merging into the undivided transverse lamellae 
of the strongly dilated discoid apex, which is without free terminal 
phalange. Body more or less depressed, dorsal lepidosis of juxtaposed, 
subuniform granules, below with smooth, subhexagonal, imbricate 
scales; tail also depressed, covered above with granules, below with 
larger imbricate scales. Pupil round; eyelid distinct all round the eye. 
Males with preano-femoral pores forming an uninterrupted series; 
both sexes with a pair of small, transverse, post-anal slits. 

* Omitted from map accompanying this paper on account of failure to locate. 

' Boettger (19i;i) emended the S[)elliiig to Phelsuniia. but this seems undesirable. Though 
named for van Phelsum. the noun has a feminine ending which makes it advisable to have all 
the adjectival specific names in agreement. 



LOVERIDGE: AFRO-ORIENTAL GECKOS 443 

Range. Coast of Tanganyika Territory ^ and islands of the Indian 
Ocean (Zanzibar; Pemba; Seychelle; Amirante; Aldabra; Comoro; 
Madagascar; Mauritius; Andaman, etc.). 





1. 2. 

Showing relative position of nostril in 

1, P. inunguis. 2. P. nmdagascariensis . 

(After Rendahl) 

Key to the Species 

I. Subcaudals subequal, the median series not strongly transversely enlarged 
A. Centre of nostril immediately above the suture between first labial 
and rostral (see fig. 1). 

1. Ventral scales smooth. 

Upper labials 8; lower labials 7; preano-femoral pores in males 
12; range : Rodriguez Island, near Mauritius newtoni 

(p. 446) 
Upper labials 8-10; lower labials 6-8; preano-femoral pores 
in males 14-25; range: R<^union Island and Mauritius 

inunguis 

(p. 448) 
Upper labials 1 1-14; lower labials 9-10; preano-femoral pores 

in males 45-46; range: Round Island, near Mauritius 

guentheri 

(p. 449) 

2. Ventral scales keeled. 

Said to agree with inunguis ( = cepediana) in all other respects 
except that dorsally the caudal verticils are composed of 7 
(instead of 8-1 1 ) scales ; range : said to have come from Mada- 
gascar trilineata 

(p. 450) 

' Boulenger (1885d, p. 214) records two young geckos from Quilimaiie, Mozambique as 
"probably introduced"; received from Sir John Kirk. 



444 bulletin: museum of comparative zoology 

B. Centre of nostril immediately above the first labial (see fig. 2). 

1 . Ventral scales keeled, even though feebly. 

a. Dorsal, or at least lumbar, granules keeled, even though 
indistinctly. 

A dark lateral line but no oval patch or ocelli in dorsolateral 
region; range: eastern Madagascar lineata 

(p. 451) 
A large vertically oval black patch above and behind fore 
limb only; range : eastern Madagascar himaculata ^ 

(p. 453) 
A large ocellus behind fore limb and another before hind 
limb; range : southcentral Madagascar quadriocellata 

(p. 454) 

b. Dorsal and lumbar granules entirely smooth. 

An orange-brown vertebral line flanked by a series of large 
orange-brown blotches; range: Seychelle Islands. . .astriata 

(p. 455) 

2. Ventral scales smooth. 

a. Posterior chin shields and lateral chest scales weakly 
keeled. 

Dark green dorsal coloring merges on flanks into whitish 
belly without marmorations; size large — 93 mm. from 
snout to anus; range: Praslin Island, Seychelle Islands 

sundbergi * 
(p. 456) 

b. Posterior chin shields and lateral chest scales as smooth as 
the ventrals. 

i. Rostral with median cleft above. 

(a) Dorsal granules entirely smooth. ' 
Dorsal pattern of dark longitudinal lines; inter-suprana- 
sal granules 1; upper labials 6-9; lower labials 6; range: 

eastern Madagascar barbouri sp. n. 

(p. 457) 
Dorsal pattern of dark cross bars; inter-supranasal 
granules 2; upper labials 9; lower labials 7; range: south- 
western Madagascar standingi 

(p. 459) 

(b). Dorsal granules obtusely keeled.* 
Dorsal pattern, if any, of small orange-brown spots; 

'Occurs with lineata from which it is possibly not distinct (Mertens), though more probably 
a synonym of quadriocellata, as suggested by Reudahl. 

' As nothing is said of the condition of the subcaudals, their subequal condition is assumed, 
should they be transversely enlarged then sundbergi will become a race of madagascarientit. 

' Much care should be used in deciding. 



loveridge: afro-oriental geckos 445 

inter-supranasal granules 2-3, rarely 1; upper labials 
9-11; lower labials 8-10; range: northwestern Madagas- 
car; Comoro Islands; Zanzibar and adjacent coast of 
Tanganyika Territory d. dubia 

(p. 460) 
ii. Rostral without median cleft above. 
Dorsal and ventral coloring usually sharply separated 
by a dark lateral line or lines; chin as far back as third 
lower labial covered with shields of which the outer are 
much larger than the inner; range: Grand Comoro 
Island ( ? above 3000 feet) d. comorensis 

(p. 463) 
Dorsal and ventral coloring intermingling on flank; chin 
as far back as the third lower labial covered with tile-like 
shields of which the outer are scarcely larger than the 
inner; range: Comoro Islands v-nigra 

(p. 464) 
II. Subcaudals with median series strongly transversely enlarged. 

A. Snout 1.6 to 2 times as long as the distance between eye and ear- 
opening. 

Back dark olive slate or brown, uniform or a narrow vertebral line 
flanked by purple streaks; range: Aldabra Islands m. abbotti 

(p. 465) 
Back green, uniform or spotted with reddish orange; range: Anda- 
man Islands m. andamanensis 

(p. 467) 
Back green or purplish, a reddish orange vertebral line flanked by 
reddish orange blotches; range: Seychelle and Amirante Islands 

tn. longinsulae 

(p. 466) 
Back bluish gray or slate, uniform or finely vermiculate with black; 
range: Pemba Island m. parkeri 

(p. 468) 
Back bluish gray or purplish, uniform or spotted with reddish 
orange, such spotting often confined to lumbar region; range: 
Madagascar and adjacent islands m. madagascariensis 

(p. 470) 

B. Snout 1.3 to 1.6 times as long as the distance between eye and ear- 
opening. 

1 . Transverse lamellae + transverse shields beneath fourth toe 
16-18; males with 24-26 (28 fide Boulenger) preano-femoral 
pores; lower labials 6-9. 

Upper labials 6-9 ; outer row of enlarged chin shields bordered 
internally by smaller ones passing gradually into gular 
granules; tail much spotted; range: western Madagascar and 



446 bulletin: museum of comparative zoology 

adjacent islands (Nosy Be; Nosy Faly), Comoro Islands (An- 
jouan; Mayotte) and Farquhar laticauda 

(p. 472) 
Upper labials 8-11; 6-10 enlarged chin shields surround a 
group of small scales; gular chevrons extend as parallel lines 
to side of neck; range: northeastern Madagascar. . . .guttata 

(p. 474) 
2. Transverse lamellae + transverse shields beneath fourth toe 
10-12; males with 27-32 preano-femoral pores; lower labials 5-6. 
Outer row of chin shields (consisting of 3-4 pairs) bordered 
posteriorly by smaller ones passing gradually into gular 
granules; range: Madagascar (chiefly southwest and south 
but recently recorded from the northeast) mutabilis 

(p. 475) 
Outer row of chin shields (consisting of only 2 pairs) bordered 
posteriorly and abruptly by minute granules (possibly not 
distinct from mutabilis or records confused); range: southern 
Madagascar brevic^ps 

(p. 477) 

Phelsuma newtonii Boulenger 

1884c. Phelsuma newtonii Boulenger, Proc. Zool. Soc. London, p. 2: Rodriguez 

Island, near Mauritius. 
1885d. Boulenger, p. 212, pi. xvii. 
1917g. Chabanaud, p. 442. 

Description. Snout 1.6 times as long as the distance between the eye 
and the ear-opening, vertical diameter of the latter less than half that 
of the former; rostral with median cleft above; centre of nostril above 
the suture between rostral and first labial; the uppermost nasal separ- 
ated from its fellow by 3 granules; upper labials 8; lower labials 7; 
only an inner pair of chin shields well differentiated from the gulars. 

Back covered with smooth granules; ventral scales smooth; males 
with 12 preano-femoral pores forming an uninterrupted series; tail, 
when unregenerate, covered above with smooth scales, below with 
smooth, imbricate, subuniform scales. 

Coloration. Above, blackish olive. Below, bluish gray; lips and 
throat sometimes whitish. 

Measurements. Total length of cotype cf , 223 (105 + 118) mm. 

Remarks. No examples seen as known only from the two cotype 
males in the British Museum and a pair in the Paris Museum. The 
digits are said to be shorter than in other species. 

Range. Rodriguez Island, near Mauritius. 



loveridge: afro-oriental geckos 



447 



^Siatistical Synopsis of Variation in ihe Genus Phelsuma 



SPECIES 


Centre of 
nostril above 
(1) rostro-labial 
suture (2) labial 


Nostril 
between first 
labial and . . . 


Number of 
inter-supra- 
nasal granules 




X ■ 

— 

^2 


Lamellae 
beneath 
fourth toe 


Range of 
prea no-femoral 
pores in males 






rostral + 












newtoni 


suture 


3 nasals 


3 


8 


7 




12 


inunguis* 
guentheri 




it 


2-3 


8-10 
11-14 


6-8 
9-10 


17-18 


14-25 
45-46 


trilineata 




l( 












lineata* 


labial 


3 nasals 


1-2 


7-10 


6-8 


15-19 


24-30 


bimaculata 




" only 


1 


8-9 


7-9 




27 


quadriocellata* 
astriata* 






1-2 
1-2-3 


9-10 
8-11 


7-9 

7-8 


16-17 
15-17 


27-29 
27-30 


sundbergi 
barbouri sp. n.* 




tt 


1-2-3 

1 


8-11 
6-9 


6-9 
6 


16 


27-36 


standingi 
d. dubia* 




ti 

" 22 


2 
1-2-3 


9 
9-12 


7 
8-10 


15-17 


? 
22-29 


d. comorensis* 




(( 


1 


7-9 


6-8 


15-16 


12-17 


v-nigra * 
m. abbotti* 




tt 

(i 


1 
1-2-3 


7-8 
7-8 


6-8 

7-7 


15-19 
18 


20-28 
33-34 


m. andamanensis* 




It 


2-3 


8-11 


6-10 


15-16 


24-32 


in. longinsulae* 




(1 


1-2-3 


9-11 


7-9 


17-18 


28-35 


m. parkeri* 




u 


1 (2) 


8-10 


7-9 


16-17 


32-38 


in. madagascariensis* 




tt 


1 


7-10 


6-9 


16-19 


34-50 


laticauda * 




tt 


1 


6-9 


6-9 


16-18 


20-28 


guttata 

mutabilis * 




tt 

2-3 nasals 


2 
1-2-3 


8-11 
6-8 


6-9 
5-6 


10-12 


27-32 


breviceps * 




2 nasals 


1 


6-7 


5-6 


11 


27-31 

c 



' Naturally many of these ranges will be extended when more material is available of those 
species of which only one or two examples are known at the present lime. 

"- Two in a Bagamoyo lizard only. 

* Represented in the collections of the Museum of Comparative Zoology, examples of those 
species without asterisk are earnestly desired. 



448 bulletin: museum of comparative zoology 

Phelsuma inunguis (Cuvier) 

1817. Gecko inunguis Cuvier, R6gne Animal, ed. 1, 2, p. 46, pi. v, fig. 3: Isle 

de France, i.e. Mauritius. 
1817. Gecko c^pMien (Peron) Cuvier, R6gne Animal, ed. 1, 2, p. 46, pi. v, 

fig. 5: Isle de France. 
1820. Gecko cepedianus Merrem, Vers. Syst. Amphibien, p. 43: Mauritius. 
1825. Phelsuma crepidanus {sic) Gray, p. 199. 
1827. Phelsuma ornatum Gray, in King, Narrative Survey inter, west. Coasts 

Australia . . .1818 and 1822, App., p. 428: Mauritius. 
1831b. Gray, p. 48. 
1831b. Phelsuma Inunguis Gray, p. 47. 
1831b. Phelsuma Cepedianus Gray, p. 47. 
1845. Gray, p. 166. 

1833. Platydactylus cepedianus Geoffrey, CI. Ill, pi. iii. 
1836. Dum6ril & Bibron, p. 301, pi. xxviii, fig. 2. 
1843. Anoplopus cepedeanus Fitzinger, p. 99. 
1877b. Pachydactylus cepedianus Peters (part), p. 455. 
1881c. Boettger (part), p. 530. 
1885d. Phelsuma cepedianum Boulenger, p. 211. 
1885c. Muller (part), p. 296. 
1887. Strauch (part), p. 17. 
1890a. Miiller, p. 290. 
1909. Mocquard, pp. 19, 96. 
1939a. Rendahl, p. 265, fig. 2a. 

Further citations of 'cepedianus' will be found under guentheri, bar- 
houri, d. dubia, m. longinsulae, m. viadagascariensis, and laticauda. 
Rochebrune's (1884a, Faune de la Senegambie. Reptiles, p. 74) record 
from Senegambia is ignored. Dumeril and Bibron's synonymy is mis- 
leading as they have latinized many vernacular citations until they 
bear little resemblance to the original. 

Description. Snout 1.7 to 2 times as long as the distance between 
the eye and the ear-opening, vertical diameter of the latter ^seven- 
eighths that of the former; rostral with median cleft above; centre of 
nostril above the suture between rostral and first labial; nostril be- 
tween rostral (sometimes barely), first labial (sometimes excluded) and 
3 nasals, the uppermost separated from its fellow by 2-3 granules; 
upper labials 8-10; lower labials 6-8; chin region covered with en- 
larged scales of which the outer are larger than the inner, the latter 
posteriorly merging into the gulars. 

Back covered with smooth granules; ventral scales smooth; males 
with 14-25 preano-femoral pores forming an uninterrupted series; 

• Halfy/cfe Boulenger (1885d). 



loveridge: afro-oriental geckos 449 

about 17-18 transverse lamellae beneath fourth toe anteriorly; tail, 
when unregenerate, covered above with smooth scales, below with 
smooth imbricate scales, subuniform, or the two median series some- 
what enlarged. 

Coloration. Above, bluish or purplish with reddish markings, viz. a 
variously-shaped interocular mark, snout with, or without, a n- 
shaped band whose arms extend to the eyes and may continue on to 
the shoulder or even along the flanks, occiput and nape with two or 
three longitudinal stripes, back and tail irregularly spotted. Below, 
whitish, the throat sometimes grayish but not spotted. 

Measurements. Total length of cf , 151 (68 + 83) mm., from Re- 
union Island (M.C.Z. 2165). 

Remarks. Gecko inunguis was placed with a query in the synonymy 
of Pachydactylus ocellatus by Boulenger, a name usually applied to a 
South African gecko now known as P. gcitje (Sparrmann, 1778). It 
seems to me, however, from Cuvier's figure and locality that it was 
the poreless female of the male lizard which he called cepedien, a name 
given to it in Peron's manuscript. Under any circumstances Cuvier is 
not the author for cepedien was first latinized by Merrem. The ques- 
tion of specific identity as between inunguis and cepediana can be 
settled only by a careful examination of the types in the Paris Museum, 
bearing in mind the extreme variability in pattern, particularly as 
between young and old of this Mauritian gecko. 

Habits. Peters (part, 1877b) states that these geckos occur in forests 
and gardens where their hard-shelled eggs may be found adhering to 
leaves. He attributes to them a uniformly high, not unpleasant note 
which may be heard in the evening and at night. Both statements re- 
quire verification for most geckos select a more stable foundation — 
such as crevices in a tree trunk — to which to attach their eggs. 

Range. Mauritius and Reunion Islands (Formerly known as lies des 
France et Bourbon). 

Phelsuma guentheri Boulenger 

1873. Platydactylus cepediana Pike (not Merrem), p. 161. 

1885c. Phelsuma cepedianum Miiller (part), p. 296 ("Mauritius")- 

1885d. Phelsuma guentheri Boulenger, Cat. Lizards Brit. Mus., 1, p. 213: 

Round Island, near Mauritius 
1887. Strauch, p. 17. 
1890a. Miiller, p. 289. 

Description. Snout nearly 2 times as long as the distance between 
the eye and the ear-opening, vertical diameter of the latter about a 



450 bulletin: museum of comparative zoology 

third that of the former; rostral with median cleft above; centre of 
nostril above the suture between rostral and first labial ; upper labials 
11-14; lower labials 9-10; chin region covered with enlarged scales of 
which the outer are slightly smaller than the inner, and well differenti- 
ated from the minute gulars. 

Back covered with smooth granules; ventral scales smooth; males 
with 45-46 preano-femoral pores forming an uninterrupted series; tail, 
when unregenerate, covered above with smooth scales, below with 
smooth, imbricate, subuniform scales. 

Coloration. Above, gray, uniform (9), or head sparsely spotted 
with black (d^) and with two subparallel black streaks from eye 
towards neck where they may terminate or continue on across nape to 
unite with their fellows from the other side to form two chevron-like 
markings. Below, whitish, throat marked with gray (Strauch). 

Mrasurrmcnts. Total length of a cotype, 223 (125 -f 98) mm., but 
surpassed by a cf , with a total length of 240 mm. (Strauch). 

Remarks. Below I give verbatim Pike's description of this lizard as 
contained in the chapter dealing with his first visit to the almost inac- 
cessible Round Island which concludes with the statement: "I believe 
this lizard is as yet undescribed." Apparently these notes deal with 
the types listed by Boulenger as received from Giinther. The last 
sentence of paragraph 3 in Pike's account would appear to have been 
added at a later date to his journal-account, perhaps on hearing from 
Giinther, but this is purely speculative, it reads: "The lizard is the 
same as the one so common in Mauritius {PJaiydaciyhis ceprdianus)," 
and should apparently have been added to the previous paragraph. 

Habits. "An active little creature found in the steep rocks on the 
mountain side. It is about 6" in length and deposits from 6 to 12 white 
eggs, the size of an ounce muslyt ball, in a row on the branches of the 
Latania glaucophylla, which are so firmly glued to the bark that they 
could not be detached without breaking." (Pike). That all the eggs 
were laid by one gecko is, of course, a mistake. 

Range. Round Island, about 25 miles from Mauritius.' 

Phelsuma trilineata Gray 

1842. Phelsuma trilineaimn Gray, Zool. Misc., p. 57: No locality given, later 

stated by Boulenger to be Madagascar. 
1885d. Boulenger, p. 212. 
1909. Mocquard, pp. 19, 93. 
1845. Phelsuma lineatum Gray (part), p. 166. 

' Muller and Strauch's four specimens are labeled Mauritius only, and may lie misidentified 
inunguis. 



loveridge: afro-oriental geckos 451 

Description. Ventrals keeled, though less strongly than in lineafa, 
while dorsally a caudal verticil consists of 7 scales. Otherwise, accord- 
ing to Boulenger, it agrees with crpediana (i.e. inunguis) in every re- 
spect. Known only from the type, allegedly from Madagascar, and a 
second specimen which is without locality, both in the British Museum, 
and neither of which I have seen. 

Coloration. Above, olive black; forehead with a triangular orange 
spot; a broad streak from nostril through ear to fore limb; three narrow 
lines and two rows of oblong orange spots. (Gray). 

Range. Said to be Madagascar. 

Phelsuma lineata Gray 

1842. Phelsuma lineatum> Gray, Zool. Misc., p. 57: Madagascar. 

1845. Gray, p. 166. 

1885d. Boulenger, p. 216. 

1885a. Miiller, p. 163. 

1887a. Boulenger, p. 490. 

1887. Strauch, p. 18. 

1893a. Boettger, p. 39. 

1901. Schenkel, p. 182. 

1901b. Tornier (part), p. 64. 

1903b. Ferreira, p. 22. 

1909. Mocquard, pp. 20, 92, 93. 

1913. Boettger (as Phelsumia), pp, 294, 322, 328. 

1913a. Methuen & Hewitt, p. 187. 

1918. Barbour, p. 482. 

1922. Kaudern, p. 420. 

1925b. Angel, p. 60. 

1931d. Angel, p. 515. 

1933b. Mertens (part), p. 265. 

1934a. Angel, p. 312. 

1939a. Rendahl, p. 270. 

1880a. Pachydactylus laticauda Boettger (part), p. 280 (Tamatave only). 

1881c. Boettger (part), pp. 461, 530 (Tamatave only). 

1881c. Pachydactylus lineatus Boettger, p. 530. 

1893a. Phelsuma laticauda Boettger (part), p. 39 (Tamatave only). 

1918. Barbour, p. 482. 

1913. Phelsumia lineata var. hifasciata Boettger, in Voeltzkow, Reise in 

Ostafrika, 3, p. 294: Ankarimbela, i.e. Ankarimbola, southcentral 

Madagascar. 
1936c. "! Phelsuma dubium AngeP (? not Boettger), p. 126 (Tsianovoka). 

' Some of the references following are to lineatus and lineata. 

- B;(sed on a young gecko; tentatively placed here as being the only record of dubia from the 
east coast, whereas P. lineata has been recorded from Anovoka by Boettger (1913). 



452 bulletin: museum of comparative zoology 

Further citations of 'lineata' will be found under trilineata, carinata, 
and bimaculata. 

Description. Snout 1.3 to 1.6 times as long as the distance between 
the eye and the ear-opening, vertical diameter of the latter about half 
that of the former; rostral without, rarely with, a median cleft above; 
centre of nostril above first labial; nostril between first labial and 3 
nasals, the uppermost separated from its fellow by 1-2 granules; upper 
labials 7-10; lower labials 6-8; chin region covered with enlarged 
scales of which the outer are slightly larger than the inner, the latter 
posteriorly merging into the gulars. 

Back covered with strongly, or obtusely, keeled granules; ventral 
scales keeled, though often very feebly; males with 24-30 preano- 
femoral pores forming an uninterrupted series; about 15-19 transverse 
lamellae beneath fourth toe anteriorly; tail, when unregenerate, 
covered above with keeled scales, below with keeled, imbricate, sub- 
uniform scales. 

Coloration. Above, brownish or purplish; snout sometimes with a 
dark, trident-like marking; a dark streak from nostril through eye and 
ear along flank and tail, sharply defining the dorsal coloring from the 
ventral; back blotched or variegated with lighter posteriorly; limbs 
and tail marbled with black. Below, whitish; a dusky violet n- 
shaped mark following contour of lower jaw to insertion of forearm, 
sometimes extending on along the flank, thereby separating a light 
lateral line {bimaculata), to hind hmb; tail immaculate or flecked with 
darker. 

Measurements. Total length of cf , 105 (53 + 52) mm., from Mada- 
gascar (U.S.N.M. 23427), and of 9 , 108 (52 + 56) mm., from be- 
tween Tamatave and Tananarivo (M.C.Z. 11719). 

Remarks. Boettger (1881c) assigned to the synonymy Dumeril & 
Bibron's (1836, p.298) Pachydactylus ocellatus on account of their 
statement as to its probable occurrence in Mauritius, though a re- 
examination of the type of Gecko ocellatus may yet demonstrate it to 
be a Phelsuma it will not be found referable to lineata. 

Boulenger (1887a), after examination of the types, referred quad- 
riocellata to the synonymy of lineata. Though it must be confessed that 
the lepidosis is closely similar I am confident that they are distinct, 
perhaps, as suggested by Mertens (1933b), subspecifically, but no 
definite locality has yet been recorded for quadriocellata. 

Boettger (1913) separated bifasciata, of which the Museum of Com- 
parative Zoology has a cotype (M.C.Z. 21950), on the grounds that 



loveridge: afro-oriental geckos 453 

it usually has 2 granules between the anterior nasals whereas lineata 
usually has 1, yet Boettger later says that of the 11 cotypes which he 
has examined 6 have 2 granules and 5 have only 1 ! The only other 
diflPerences cited are the allegedly less well defined caudal verticils, a 
character which appears variable in most species of the genus, and the 
presence of a second longitudinal stripe on the flank. Yet in our cotype 
the stripe does not extend behind the fore arm and appears to be repre- 
sented in varying degrees of indistinctness in our extensive material 
from other localities. 

Localities. Madagascar: Alaotra; Ambatondrazonaka; Ambila; 
Ambilo; Ambinany (? Eminiminy); Anamalozotra ; Ankarimbola 
(Ankarimbela) ; Anovoka (Anevoka; Tsianovoka); Antsihanaka; 
Betampona Reserve; (Eminiminy, see Ambinany); Didy Swamps; 
Fandrarazana ; Fenerive; Fianarantsoa; lie aux Prunes; Ivoloina;- 
(Majunga^); Sakana; Ste. Marie Id. Tamatave; Tamatave to Tanan 
arivo; Tampina. 

Mertens (1933b) record from Col Pierre Radama is removed to 
himaculata until the status of that form is definitely settled. 

Range. Eastern^ Madagascar. 

Phelsuma bimaculata Kaudern 

1922. Phelsuma bimaculata Kaudern, Zool. Jahrb. Syst., 45, p. 420, pi. xii, 

fig. 2: Fandrarazana, eastern Madagascar. 
1939a. Rendahl, p. 270. 
1933b. Phelsuma lineata Mertens (part), p. 265. 

Description. Snout (1.6 to) 2 times as long as the distance between 
the eye and the ear-opening, vertical diameter of the latter about half 
that of the former; rostral with, rarely without, median cleft above; 
centre of nostril above first labial; the uppermost nasal separated from 
its fellow by 1 granule; upper labials 8-9 (11); lower labials (6) 7-9. 
Back covered with slightly keeled granules; ventral scales strongly 
keeled; males with 27 (24-28) preano-femoral pores forming an unin- 
terrupted series; tail, when unregenerate, covered above with keeled 
scales. 

Coloration. Above, bluish or reddish violet flecked with golden 
yellow; above and behind fore limb is a large, vertically oval black 
blotch (but no lateral line as is found in lineata); Hmbs shghtly mottled; 
tail punctate with black dorsally and with an indistinct, longitudinal, 

' Boettger's (1893a, 1913) record of Majunga, being the only one from the northwest should 
be received with reserve, the possibility of erroneous locality data not being overlooked. 



454 bulletin: museum of comparative zoology 

black line laterally. Below, whitish, a dusky violet n -shaped mark fol- 
lowing contour of lower jaw to insertion of fore Hmh; tail immaculate. 

Rendahl (1939a) says that Kaudern overlooked a faint second 
blotch anterior to hind limb in the types. Kaudern is mistaken in 
thinking that the underside of the tail in all lineata are punctate, some 
of our numerous specimens have tails which are immaculate beneath. 

Measurements. Total length of cotype cf , 85 (43.5 + 42.5) mm., of 
cotype 9 , 89 (43.5 + 45.5) mm., but tails of both partly or wholly 
regenerated. 

Remarks. Mertens (1933b), who had nine geckos from Col Pierre 
Radama which agreed with the figure and description of himaculata, 
is inclined to think that it is but a color variant of lineata for the latter 
species was taken in the same locality. Kaudern also referred to 
lineata six geckos which he took at Fandrarazana, the type locality of 
himaculata. 

Rendahl (1939a) compared the types of himaculata with six geckos, 
taken in southeast Madagascar between the tableland and coast, and 
declared them to be the same though his specimens exhibited four 
blotches as in quadriocellata, a species with which he thought himacu- 
lata might prove to be synonymous, a conclusion which is probably 
correct. In view, however, of the possibility that they are subspeci- 
fically distinct, and that his material from the southeast should rather 
be referred to quadriocellata of southcentral Madagascar than to 
himaculata of northeastern Madagascar, any additional data derived 
from Rendahl's six specimens is incorporated in the above description 
in parentheses. 

Localities. Madagascar: Col. Pierre Radama; Fandrarazana. 

Range. Eastern Madagascar. 



Phelsuma quadriocellata (Peters) 

1883b. Pachydactylus quadriocellatus Peters, Sitz. Ges. Naturf. Freunde Berlin, 

p. 28: Madagascar. 
1885d. Phelsuma quadriocellatns Boulenger, p. 216. 
1909. Phelsuma quadriocellatum Mocquard, pp. 20, 98. 

Description. Snout 1.5 times as long as the distance between the 
eye and the ear-opening, vertical diameter of the latter about half that 
of the former; rostral without median cleft above; centre of nostril 
above first labial; nostril between first labial and 3 nasals, the upper- 
most separated from its fellow by 1-2 granules; upper labials 9-10; 
lower labials 7-9; chin region covered with enlarged scales of which the 



loveridge: afro-oriental geckos 455 

outer are but slightly larger than the inner, the latter posteriorly 
merging into the gulars. 

Back covered with almost smooth granules; ventral scales keeled; 
males with 27-29 preano-femoral pores forming an uninterrupted 
series; about 16-17 transverse lamellae beneath fourth toe anteriorly; 
tail, when unregenerate, covered above with smooth scales, below with 
smooth (distally) and keeled (basically) imbricate, subuniform scales. 

Coloration. Above, emerald green, more or less uniform or flecked 
with darker; head and neck spotted with pale blue; behind the fore 
limb is a black spot edged with pale blue, a similar spot in front of the 
hind limb; on the flanks the dorsal pigmentation darkens along its 
lower edge (as in Imeata). Below, whitish (yellowish), a dusky violet 
n -shaped mark following contour of lower jaw to insertion of fore limb, 
sometimes extending on along the flank to hind limb. Iris golden 
yellow. 

Measurements. Total length of cT, 88 (42 + 46) mm.; of 9 , 80 
(38 + 42) mm. (M.C.Z. 7734). 

Remarks. Despite its close relationship to lineata and the probability 
that the ocelli are derived from a dark lateral line, I cannot help but 
think, from examination of the pair in the Museum of Comparative 
Zoology, that this rare lizard is distinct; possibly, as suggested by Mer- 
tens (1933b), it maj' ultimately prove to be a geographical form when 
we know something of its distribution. 

Range. Central Madagascar. 



Phelsuma astriata Tornier 

1901b. Phelsuma lineata var. astriata, Tornier, Zool. Anz., 24, p. 65: Mahe 

Island, Seychelle Islands. 
1939a. Phelsuma carinatum Rendahl, Zool. Jahrb. Syst., 72, p. 277: Praslin 

Island, Seychelle Islands. 
1939a. Phelsuma carinatum Tnaheense Rendahl, Zool. Jahrb. Syst., 72, p. 278: 

Port Victoria, Mahe Island, Seychelle Islands. 

Description. Snout 1.8 to 2.1 times as long as the distance between 
the eye and the ear-opening, vertical diameter of the latter about half 
that of the former; rostral with median cleft above; centre of nostril 
above first labial; nostril between first labial and 3 nasals, the upper- 
most separated from its fellow by 1-3 granules; upper labials 8-11; 
lower labials 7-8 ; chin region covered with enlarged scales of which the 
outer are larger than the inner, the latter posteriorly merging into the 
gulars. 



456 bulletin: museum of comparative zoology 

Back covered with smooth granules; ventral scales keeled; males 
with 27-30 preano-femoral pores forming an uninterrupted series; 
about 15-17 transverse lamellae beneath fourth toe anteriorly; tail, 
when unregenerate, covered with smooth scales, below with keeled, 
imbricate, subuniform scales (in reproduced tails sometimes trans- 
versley enlarged, sometimes not). 

Coloration. Above, blue violet; a brownish-orange streak from nos- 
tril to eye; a n-shaped mark in frontal region; back with a vertebral 
line (sometimes broken up) flanked by more or less coalescing spots; 
flanks with another series; limbs variegated with lighter; tail spotted 
or cross-barred with lighter. Below, whitish, a dusky violet n-shaped 
mark following contour of lower jaw; breast, belly, and tail immacu- 
late. 

The coloring closely resembles that of P. m. longinsulae, also of the 
Seychelles. It is interesting to note that Rendahl says of carinatum 
that the coloring resembles that of sundhergi but vertebral stripe is 
darker and the red markings larger, while of maheense he remarks that 
it resembles carinatum except for the more distinct gray vertebral 
stripe. 

Measurements. Total length of d", 123 (57+66) mm. (M.C.Z. 
46146), largest cotype of carinatum 56 mm. from snout to vent; of 
maheense 55.5 mm. ; tails of both being damaged. 

Remarks. No characters or key for distinguishing carinatum and 
maheense are furnished by Rendahl who apparently overlooked 
Tornier's astriata, buried in the text under the heading Phelsuma 
lineata. 

Localities. Seychelle Islands: Bel Air, near Port Victoria, Mahe 
Id.; Praslin Id. 

Range. Seychelle Islands. 

Phelsuma sundbergi Rendahl 

1939a. Phelsuma sundhergi Rendahl, Zool. Jahrb. Syst., 72, pp. 274, 318, figs. 
3-5, 16: Praslin Island, Seychelle Islands. 

Description.. Snout about 2.1 times as long as the distance between 
the eye and the ear-opening; rostral with median cleft above; centre 
of nostril above first labial; nostril between first labial and 3 nasals, 
the uppermost separated from its fellow by 1-3 granules; upper labials 
8-11; lower labials 6-9. Posterior chin shields and lateral scales of 
chest weakly keeled; ventral scales smooth; males with 27-36 preano- 
femoral pores forming an uninterrupted series. 



loveridge: afro-oriental geckos 457 

Coloration. Above, dark green (almost emerald green in life); a 
reddish streak from nostril to eye present or absent; a n -shaped mark 
on frontal region; on flanks the dark dorsal coloring merges into the 
lighter coloring of the belly without marm orations ; limbs uniform 
(weakly marbled in life). Below, whitish; a dusky n-shaped mark 
following contour of lower jaw present or absent; throat of young 
uniform or shaded with gray, of adults with dusky flecking. Tongue 
red. 

Measurements. Total length of largest cotype, 178 (93.5+84.5) mm. 

Remarks. As nothing is said about the subcaudal arrangement, the 
present action in placing sundbergi in the group of those having sub- 
uniform subcaudals is arbitrary and based on Rendahl's action in 
placing it before "carinatum." Should the subcaudals have the 
median series transversely enlarged, sundbergi will undoubtedly prove 
to be a race of mascareniensis (which has been recorded from Praslin 
Id. by other authors) possibly peculiar to Praslin where the keeling 
on posterior chin shields and breast may have developed as a secondary 
sexual character for one notes that no fewer than 17 of the 24 cotypes 
are males. Had it not been for Rendahl's mentioning young speci- 
mens in his color description I should have been tempted to suggest 
uniting sundberg with longinsulae (the former having page precedence) 
whose head markings it is said to resemble. 

Diet. In captivity, flies, cockroaches, and mealworms (Rendahl). 

Habits. In captivity these geckos were eager to take up drops of 
water with their tongues. Though diurnal, they avoided direct sunlight 
and sought out shady corners, preferring the glass sides of their vivar- 
ium to the twigs and branches which had been supplied them. 
Occasionally one male might be seen chasing another but at no time 
was any sound heard from them. As the sloughing period approached, 
the geckos' movements became perceptibly slower and a striking 
change from the normal bright green coloring to dusky blackish green 
took place. As the old skin loosened it was removed and eaten by the 
gecko. After a year in captivity a skin disease made its appearance 
among them, it was characterized by black spots and wasting of the 
tail, which would finally fall off. Assuming it to be a deficiency disease, 
Vigantol on sugar was supplied with some success, for the last geckos 
to succomb were not victims of the disease. 

Range. Praslin Island, Seychelle Islands. 



458 bulletin: museum of comparative zoology 

Phelsuma barbouri spec. nov. 
1918. Phelsuma cepedianum Barbour (not Merrem), p. 482. 

Type. Museum of Comparative Zoology, No. 11640, a 9 from forest 
between Tamatave and Tananarivo, eastern Madagascar, collected by 
F. R. Wulsin, 1915. 

Faratypes. Museum of Comparative Zoology, No. 11641, a younger 
9 with same data as type; also American Museum of Natural History, 
No. 47897, an adult 9 from Madagascar (precise locality label dis- 
integrated by preservative), collected by the Archbold Franco-Anglo- 
American Expedition of 1929-1931. 

Diagnosis. Closely related to stmidingi Methuen & Hewitt, of 
southwestern Madagascar, from which it differs in those characters 
indicated in the Key (p. 444) and to which it may be subspecifically 
related but the males of neither species are known. 

Description. Snout 1.3 to 1.5 times as long as the distance between 
the eye and the ear-opening, vertical diameter of the latter about 
three-quarters that of the former; rostral with median cleft above; 
centre of nostril above first labial; nostril between first labial and 3 
nasals, the uppermost separated from its fellow by 1 granule; upper 
labials 6-9; lower labials 6; anterior pair of chin shields much larger 
than those immediately behind, the latter posteriorly merging into 
the gulars. 

Back covered with smooth granules; ventral scales smooth; (male 
unknown) ; about 16 transverse lamellae beneath fourth toe anteriorly; 
tail, when unregenerate, covered above with smooth scales, below 
with smooth, imbricate, subuniform scales. 

Coloration. Above, brownish olive; a dark brown streak from 
nostril through eye and ear-opening to flank anteriorly; another 
streak from supraocular region continuing as a dorsolateral line to 
about midbody where it breaks up or continues indistinctly to base of 
tail; two brown streaks from snout to occiput where they coalesce to 
form a single line on nape only to break up into a series of dashes 
posteriorly, in fact on the dorsum there is a tendency for all these 
lines to disintegrate and form vermiculations; limbs almost uniform; 
tail uniform. Below, uniformly bluish gray except for an obsolescent, 
dusky, n-shaped mark following contour of lower jaw. 

Measurements. Total length of type 9 , 99 + (59 -|- 40+) mm., the 
tip being regenerated, surpassed in length from snout to anus by a 
paratype 9 (A.M.N.H. 47897) of 64 mm. 

Remarks. At the time these two specimens were received the genus 



loveridge: afro-oriental geckos 459 

was but poorly represented in the collection of the Museum of Com- 
parative Zoology with less than half the forms, we have today. 

Phelsuma STANDING! Methuen & Hewitt 

1913a. Phelsuma standingi Methuen & Hewitt, Ann. Transvaal Mus. 3, p., 
187: Forest fringing Onilahy River at Maroamalona, southwestern 
Madagascar. 

1939a. Rendahl, p. 268. 

Description. Snout 1.5 times as long as the distance between the 
eye and the ear-opening; rostral with median cleft above; nostril 
between first labial and 3 nasals, the uppermost separated from its 
fellow by 2 granules; upper labials 9; lower labials 7; anterior pair of 
chin shields much larger than those immediately behind. 

Back covered with smooth granules; ventral scales smooth; (male 
unknown); tail, when unregenerate, covered above with smooth 
scales, below with smooth, imbricate scales (of which the two median 
rows are enlarged, according to the authors). 

Coloration. Above, emerald green, with numerous dark, irregular, 
transverse bars. Below, whitish, except for some bluish-gray flecking 
on throat; tail bluish. 

Measurements. Total length of type 9, 149+ (88 + 61+) mm., 
the tail being regenerated. 

Remarks. Known to me only from the scanty description which 
should be augmented to the complete format furnished for other 
species. I am indebted to Mr. V. I'itzSimons for confirming points in 
the description in which it differed from the preceding species. 

Locality. Madagascar: Maroamalona. 

Range. Southwestern Madagascar. 



460 bulletin: museum of comparative zoology 



Phelsuma dubia dubia (Boettger) 

1854. Platydactylus cepedianus Peters (not Merrem), p. 615. 

1855. Peters, p. 44. 
1887. Vaillant, p. 134. 

1866b. Pachydactylus cepedianus Peters (not Merrem), p. 887. 

1869a. Peters (part), p. 13 (omit Nosy Be and Seychelles). 

1882a. Peters (part), p. 27. 

1881a. Pachydactylus dubius Boettger, Zool, Anz., 4, p. 46: Madagascar. 

1881c. Boettger, pp. 464, 530. 

1881g. Boettger, p. 179. 

1885d. Phelsuma madagascariense Boulenger (part), p. 214 (Zungomero). 

1885d. Phelsuma dubius^ Boulenger, p. 215 (no material). 

1893a. Boettger, p. 38. 

1894e. Boulenger, p. 723. 

1902b. Tornier, p. 581. 

1909. Mocquard, pp. 20, 89. 

1913. Boettger^, pp. 292, 327, 336, 339, 341, 343, pi. xxix, figs. 7-10. 

1913c. Nieden, p. 68. 

1917. Voeltzkow, p. 460. 

1922a. Mertens, p. 183. 

1931d. Angel, p. 515. 

1939a. Rendahl, p. 268. 

1940. Parker, Moreau & Pakenham, p. 309. 

1941. Moreau & Pakenham, 1940, p. 107. 

1900b. Phelsuma laticauda Tornier (not Boettger), p. 588. 

1913a. Methuen & Hewitt, p. 187. 

1920a. Loveridge, p. 139. 

1923d. Loveridge, p. 846. 

1924b. Loveridge, p. 9. 

1925a. Loveridge, p. 72. 

1928c. Barbour & Loveridge, p. 146. 

1933h. Loveridge, p. 295. 

1937f. Loveridge, p. 492. 

1903b. Phelsuma cepedianum Ferreira (not Merrem), p. 21. 

1917. Voeltzkow, p. 459. 

Further citations of 'dubia' will be found under lineata and laticauda. 

Native name. Camantindi (Anjouan Id.: Peters). 

Description. Snout 1.3 to L6 times as long as the distance between 
the eye and the ear-opening; vertical diameter of the latter about 
three-quarters that of the former; rostral with^ median cleft above; 
centre of nostril above first labial; nostril between first labial and 2-3 

' Some of the references following are to dubium and dubia. 

2 As Phelsumia (1913 only). 

3 Dar es Salaam specimens allegedly without, not available for reexamination. 



loveridge: afro-oriental geckos 461 

nasals S the uppermost separated from its fellow by 2-3, very rarely 1, 
granules; upper labials 9-12; lower labials 8-10; chin region covered 
with enlarged scales of which the outer are much larger than the inner, 
the latter posteriorly merging into the gulars. 

Back covered with obtusely keeled' granules; ventral scales smooth; 
males with 19-29^ preano-femoral pores forming an uninterrupted 
series; about 15-17 transverse lamellae beneath fourth toe anteriorly; 
tail, when unregenerate, covered above with obtusely keeled scales, 
below with smooth, imbricate, subuniform scales. 

Coloration. Above, bluish gray to purplish brown, variegated and 
spotted with orange brown; on flank a dusky streak is present or 
absent; limbs vermiculated or spotted with black, gray, or bluish; tail 
sometimes bluish. Below, whitish, immaculate, or a dusky n-shaped 
mark following contour of lower jaw to shoulder. 

In life (Dar es Salaam). Above, dark green, finely flecked with red 
on back and tail. 

Judging by a hatchling (A.M.N.H. 147880), 21 mm. in length from 
snout to anus, the young are minutely speckled with brown-edged 
white spots which disappear by the time the gecko attains 23 or 24 mm. 
(A.M.N.H. 47881-2) in length. 

From time to time Boettger has cited color differences allegedly 
distinguishing duhia from other species; these do not appear to hold 
good in series and should be accepted with reserve. 

Measurements. Total length of cf , 150 (65 + 85) mm., from Zanzi- 
bar (M.C.Z. 19123); and of 9 , 133 (58 + 75) mm., from Madagascar 
(Boettger, 1913). 

Remarks. The type of dubia was not available to Boettger when he 
(1881c) identified a specimen, allegedly from Nosy Be, as dubia. 
Later (1913) he denied the occurrence of dubia on Nosy Be 'as for- 
merly supposed' though I am not clear whether it was the determina- 
tion or the locality data which was at fault. Nosy Be is, however, 
omitted from the list of localities below. 

In the latter paper (1913, p. 337) he gives a very fair summary of 
the characters distinguishing dubia from laticauda, though in stating 
that the two species agree in not having the median series of sub- 
caudals enlarged, he errs, for of course laticauda has. At some time 
Boettger appears to have confused the type which he originally 
stated was in the Hamburg Museum and later referred to as being in 

• Two is rare, in one Bagamoyo (M. C. Z. 24084) gecko only. 

2 Smooth, in the original description, was subsequently corrected. 

' 19-22 in Methuen & Hewitt's (1913a) series of "laticauda," fide FitzSimons (1942. in letter) 
who says that "the reduction in pores has taken place distally, where there are 3-4 enlarged 
poreless scales continuous with those bearing pores." 



462 bulletin: museum of comparative zoology 

the Senckenberg Museum (No. 4192.2a), which Mertens (1922a) 
refutes. 

In referring Zanzibar geckos to laticauda, I (1920a, et seq.) was in 
error, having accepted Tornier's (1900b) conclusions and for several 
years overlooked his subsequent (1902b) retraction. Even then, in 
the hope that they might prove to be subspecifically related, I con- 
tinued to use the name laticauda while awaiting an opportunity to 
revise the genus. Parker (1940) recently pointed out my error and 
now, after careful comparison of Zanzibar, Bagamoyo, and Dar es 
Salaam material with a series of dubia from Madagascar, I entirely 
fail to find any difference between then, beyond, perhaps an average 
in size. 

Breeding. Presumably the pair of eggs are deposited in the crowns 
of coconut palms, for the lowest eggshells which I have found were 
at a height of six feet from the ground. 

Diet. Ants and beetles. 

Parasites. Red acarine parasites are sometimes present about anus. 

Enemies. One was recovered from the stomach of a tiger snake 
( Tarhophis s. semiannulatus) at Bagamoyo. 

Habitat. In East Africa, owing to its dwelling in the crowns of 
coconut palms, this gecko is exceedingly difficult to obtain. The first 
specimens which I secured were taken while passing through a street 
in the native quarter of Dar es Salaam. Overhead some thatchers 
were cutting fronds and dropping them into the road. Even then, one 
of the geckos made a dash for the nearby trunk of a palm which it 
ascended to safety. Occasionally one encounters these beautiful 
geckos on the ground, presumably en route from one palm to another. 

In Madagascar, according to Boettger (1913), this species is found 
upon the satra palms, whereas madagascariensis, though inhabiting 
the same localities, dwells among the mangroves. 

Localities. Madagascar^: Kandrany (Kandani); Majunga; Mak- 
amby Island; Maroamalona; St. Augustin; Soalala. Comoro Islands: 
Anjouan (Johanna) Id.; Grand Comoro Id.-; Mayotte Id.; Moheli Id. 
Zanzibar Island, Tanganyika Territory: Bagamoyo; Dar es 
Salaam; Singino; Zungomero. 

Range. Northwest Madagascar^; Comoro Islands; Zanzibar Island; 
and adjacent coast of Tanganyika Territory. 

'The only record from eastern Madagascar for dubia is that of Angel (1936c) for a young 
gecko from Anovoka (Tsiaiiovoka). I have omitted this as questionable and as the point can- 
not be cleared up at the present time by an examination of the specimen in question, have 
tentatively referred the reference to lineala, a species which has been recorded from Anovoka 
by Boettger (1913). 

2 A possible explanation of the occurrence of both the typical form and the race comorensis 
on this island may be found that the latter is at present known only from above 3000 feet- 



loveridge: afro-oriental geckos 463 



Phelsuma dubia comorensis Boettger 

1913. Phelsimiia laticauda var. comorensis Boettger, in Voeltzkow, Raise in 
Ostafrika, 3, p. 336: La Grille, circa 1000 metres, Grand Comoro 
Island. 

1917. Voeltzkow, p. 460. 

1929d. Barbour & Loveridge, p. 314. 

1939a. Rendahl, p. 269. 

Description. Snout 1.5 to L6 times as long as the distance between 
the eye and the ear-opening, vertical diameter of the latter about 
three-quarters that of the former; rostral without median cleft above; 
centre of nostril above first labial; nostril between first labial and 
3 nasals, the uppermost separated from its fellow by 1 granule (in all 
23 cotypes) ; upper labials 7-9 ; lower labials 6-8 ; chin region covered 
with enlarged scales of which the outer are much larger than the inner, 
the latter posteriorly merging into the gulars. 

Back covered with smooth granules; ventral scales smooth; males 
with 12-17^ prea no-femoral pores forming an uninterrupted series; 
about 15-16 transverse lamellae beneath fourth toe anteriorly; tail, 
when unregenerate, covered above with smooth scales, below with 
smooth, imbricate, subuniform scales. 

Coloration. Above, olive green, greenish or violet gray; a light 
streak from nostril to eye; a n-shaped mark in frontal region; a 
semicircular patch on occiput, all three light markings showing a 
tendency to coalesce; back with, or without, a light, irregular, verte- 
bral line (sometimes disintegrating) flanked posteriorly by a series of 
light irregular blotches which tend to coalesce; sometimes a broad, 
dark line along flank from eye to base of tail sharply defining the dorsal 
coloring from the ventral. Below, whitish, immaculate, or a dusky, 
violet n-shaped mark following contour of lower jaw to insertion of 
fore arm, sometimes extending on along the flank, thereby separating 
a light lateral line, to hind limb ; tail immaculate or flecked with darker. 

Measurements. Total length of cotype 9 , 85 + (50 + 35 +) mm. 
(M.C.Z. 21954). 

Remarks. The above description is based solely on three female 
cotypes in the Museum of Comparative Zoology together with data 
culled from the original description of the twenty-three cotypes. 

Boettger made comorensis a race of laticauda on account of its 
agreeing with that species in its rostral being without a median cleft 

' Perhaps Boettger intended to qualify tins by saying 'on either side,' in that event 24-34 
would seem to be extremely high. 



464 bulletin: museum of comparative zoology 

and the uppermost nasal being separated from its fellow by a single 
granule, he disregarded the more important character of its subuniform 
subcaudals as contrasted with those of laticauda in which those of 
the median series are transversely enlarged. To me its derivation 
from dubia seems obvious. We are, however, faced with the fact that 
both forms occur on Grand Comoro and can only speculate that while 
dubia is restricted to the lowlands, comorensis is a montane race. 



Phelsuma v-nigra Boettger 

1913. Phelsumia v-nigra Boettger, in Voeltzkow, Raise in Ostafrika, 3, pp. 

337, 339, 341, 343, pi. xxv, fig. 9: Moheli Island, Comoro Islands. 
1917. Voeltzkow, p. 460. 
1929d. Barbour & Loveridge, p. 314. 
1939a. Rendahl, p. 267. 

Description. Snout 1.5 to 1.6 times as long as the distance between 
the eye and the ear-opening, vertical diameter of the latter about half 
that of the former; rostral without median cleft above; centre of 
nostril above first labial; nostril between first labial and 3 nasals, the 
uppermost separated from its fellow by 1 granule; upper labials 7-8; 
lower labials 6-8; chin region covered with enlarged tile-like, flat 
scales of which the outer are slightly larger than the inner, the latter 
posteriorly merging into the gulars. 

Back covered with smooth granules; ventral scales smooth; males 
with 20-28 preano-femoral pores forming an uninterrupted series; 
about 15-19 transverse lamellae beneath fourth toe anteriorly; tail, 
when unregenerate, covered above with smooth scales, below with 
smooth imbricate scales of which the two median series are irregularly 
and slightly enlarged. 

Coloration. Above, purplish brown (blue green in life),. uniform or 
flecked with lighter gray green or gray. brown, head with from one to 
seven flecks; tail vermiculated with black. Below, whitish (citron 
yellow), a dusky n-shaped mark following contour of lower jaw; 
throat uniform or with a second dark, chevron-like mark. 

Measurements. Total length of cotype cf, 95 (45 + 50) mm. 
(A.M.N.H. 24772); of cotype 9, 94+ (46 + 48+) mm., tail regen- 
erating (M.C.Z. 17844); both from Moheli Island, which, it may be 
noted, is the type locality, not Grand Comoro Island. 

Remarks. In scale counts indistinguishable from P. d. comorensis 
of La Grille Mountain, Grand Comoro, but very different in chin 
shields and coloration. In the latter character it reveals -affinities 



loveridge: afro-oriental geckos 465 

with P. m. ahhotti of Aldabra Island, displaying the dusky gular 
chevron so frequently present in members of the madagascariensis 
group, though it differs from them in having the two median rows of 
subeaudals subequal or but slightly enlarged. Boettger observes that 
the chin shields of v-nigra are more numerous than in madagascariensis. 

Localities. Comoro Islands: Anjouan (Johanna) Id.; Mayotte Id.; 
Moheli Id. ; and Mt. Msotzo, circa 300 metres, Grand Comoro Id. 

Range. Comoro Islands. 



Phelsuma madagascariensis abbotti Stejneger 

1893b. Phelsuma ahhotti Stejneger, Proc. U. S. Nat. Mus., 16, p. 716: Aldabra 

Island. 
1893. Abbott, p. 762. 
1939a. Rendahl, p. 266. 

1909h. Phelsuma madagascariense Boulenger (part), p. 297 (Aldabra). 
191 Id. Phelsuma madagascariense var. ahhotti Boulenger, p. 378. 
1913. Boettger, p. 333. 

Description. Snout 1.6 times as long as the distance between the 
eye and the ear-opening, vertical diameter of the latter not half that 
of the former; rostral with median cleft above; centre o